New Philosophical

Essays on
Love and Loving
Edited by Simon Cushing
New Philosophical Essays on Love and Loving
 Simon Cushing
Editor

New Philosophical
Essays on Love and
Loving
Editor
Simon Cushing
University of Michigan–Flint
Flint, MI, USA

ISBN 978-3-030-72323-1    ISBN 978-3-030-72324-8 (eBook)

https://doi.org/10.1007/978-3-030-72324-8

© The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer
Nature Switzerland AG 2021
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights of
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information
in this book are believed to be true and accurate at the date of publication. Neither the
publisher nor the authors or the editors give a warranty, expressed or implied, with respect to
the material contained herein or for any errors or omissions that may have been made. The
publisher remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Cover illustration: © Sergio Amiti / Getty Images

This Palgrave Macmillan imprint is published by the registered company Springer Nature
Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Thomas and Frederick, because some love is simple and obvious. Jami too,
even though she doesn’t believe in the stuff
 Contents

1 Introduction  1
Simon Cushing

2 Making Room for Love in Kantian Ethics 25

Ernesto V. Garcia

3 Iris Murdoch and the Epistemic Significance of Love 39

Cathy Mason

4 ‘Love’ as a Practice: Looking at Real People 63

Lotte Spreeuwenberg

5 Love, Choice, and Taking Responsibility 87

Christopher Cowley

6 Not All’s Fair in Love and War: Toward Just Love Theory101
Andrew Sneddon

7 Doubting Love125
Larry A. Herzberg

8 Love and Free Agency151

Ishtiyaque Haji

vii
viii CONTENTS

9 Sentimental Reasons171
Edgar Phillips

10 Wouldn’t It Be Nice: Enticing Reasons for Love195

N. L. Engel-Hawbecker

11 Love, Motivation, and Reasons: The Case of the

Drowning Wife215
Monica Roland

12 Can Our Beloved Pets Love Us Back?241

Ryan Stringer

13 Romantic Love Between Humans and AIs: A Feminist

Ethical Critique269
Andrea Klonschinski and Michael Kühler

14 Patriotism and Nationalism as Two Distinct Ways of

Loving One’s Country293
Maria Ioannou, Martijn Boot, Ryan Wittingslow, and
Adriana Mattos

Index315
 Notes on Contributors

Martijn Boot is an assistant professor at the University of Groningen.

Prior to coming to Groningen, he was an associate professor at Waseda
University in Tokyo, Japan. He holds a DPhil from the University of
Oxford. He specializes in political philosophy. His book, titled
Incommensurability and Its Implications for Practical Reasoning, Ethics
and Justice (2017), is the result of research on incommensurability of val-
ues and its implications for justice.
Christopher Cowley is an associate professor in the School of Philosophy,
University College Dublin, Ireland. He is the author of two books (includ-
ing Moral Responsibility, 2013) and has edited three volumes (including
The Philosophy of Autobiography, 2015, and Supererogation, 2015). He
specializes mainly in ethics, with a side interest in the philosophy of crim-
inal law.
Simon Cushing is Associate Professor of Philosophy at The University of
Michigan–Flint. He co-edited The Philosophy of Autism (2012) with Jami
L. Anderson, and edited Heaven and Philosophy (2017). His interviews
with leading philosophers can be found at https://www.cognethic.org/
philosophical-­profiles.
N. L. Engel-Hawbecker is a doctoral candidate at The University of
Texas at Austin.
Ernesto V. Garcia is an assistant professor in the Philosophy Department
at the University of Massachusetts Amherst. He works mainly on history
of modern philosophy (especially Hume and Kant) and contemporary

ix
x NOTES ON CONTRIBUTORS

moral and political philosophy, and he has published on various topics

including forgiveness, authenticity, and respect for persons.
Ishtiyaque Haji is Professor of Philosophy at the University of Calgary.
He has research interests in ethical theory, philosophy of action, meta-
physics, and philosophical psychology. He is the author of Moral
Appraisability (1998), Deontic Morality and Control (2002), Moral
Responsibility, Authenticity, and Education (with Stefaan Cuypers;
2008), Freedom and Value (2009), Incompatibilism’s Allure (2009),
Reason’s Debt to Freedom (2012), Luck’s Mischief (2016), and The
Obligation Dilemma (2019).
Larry A. Herzberg is Professor of Philosophy at the University of
Wisconsin, Oshkosh, where he teaches courses on epistemology, the phi-
losophy of emotion, and the philosophy of language (among others).
Since earning his PhD in philosophy from UCLA in 2003, he has pub-
lished articles focusing on issues in the philosophy of emotion and the
epistemology of self-knowledge, including “Constitutivism, Belief and
Emotion” (2008), “Causation, Direction, and Appraisal Theories of
Emotion” (2009), “To Blend or to Compose: A Debate About Emotion
Structure” (2012), “On Knowing How I Feel About That: A Process-­
Reliabilist Analysis” (2016), “Can Emotional Feelings Represent
Significant Relations?” (2018), and “On Sexual Lust as an Emotion”
(2019). He is currently exploring epistemological issues related to the
internet.
Maria Ioannou has been an assistant professor at the University of
Groningen since 2016. She holds a DPhil from the University of Oxford
with a specialization in social psychology. Her research lies in the field of
intergroup conflict, prejudice, and intergroup contact. Prior to her
appointment at the University of Groningen, she worked as a post-
doctoral researcher at the University of Cyprus as well as a researcher
in the civil society sector conducting research on social cohesion and
reconciliation.
Andrea Klonschinski is a postdoctoral researcher at the Department
of Medical Ethics and History of Medicine, Göttingen University
(Germany). She works in the field of moral philosophy and philosophy of
economics. Her research focuses especially on the ethics of discrimination,
well-being measures, and on women in philosophy. She is the author of
the book The Economics of Resource Allocation in Health Care: Cost-Utility,
 NOTES ON CONTRIBUTORS xi

Social Value, and Fairness (2016). From 2012 until 2020, she organized
the philosophical movie series Filmisches Philosophieren, in which philo-
sophical aspects of popular movies are discussed with the audience.
Michael Kühler is a Research and Teaching Fellow at the Academy for
Responsible Research, Teaching, and Innovation (ARRTI) at Karlsruhe
Institute of Technology (KIT), Germany, as well as “Privatdozent”
(roughly equaling associate professor) at the University of Münster,
Germany. He has studied philosophy, musicology, and pedagogy and got
his doctoral degree in philosophy at the University of Erlangen-Nürnberg,
Germany, in 2005, with a thesis on the intertwined problem of moral jus-
tification and motivation. In 2012, he completed his “habilitation” at the
University of Münster, Germany, with a monograph on the principle
“ought implies can.” His areas of expertise include ethics, metaethics,
political philosophy, and the philosophy of love. He has several publica-
tions on the philosophy of love. Most recently, he co-edited (together
with Rachel Fedock and Raja Rosenhagen) the volume Love, Justice, and
Autonomy, Routledge, 2021.
Cathy Mason is Stipendiary Lecturer in Philosophy at Wadham College,
University of Oxford, supervising students in ethics, practical ethics, the-
ory of politics, and aesthetics. Her research is focused on ethics, episte-
mology (especially moral epistemology), aesthetics, and the writing of Iris
Murdoch (particularly at the points in her work where these topics con-
verge). In the summer of 2019, she completed her PhD in Philosophy,
titled Neglected Virtues: Love, Hope, and Humility, at Trinity College,
Cambridge, under the supervision of Paulina Sliwa and Tom Dougherty.
Adriana Mattos is Lecturer in Health and Life Sciences at the University
of Groningen. She holds a PhD in Medical and Pharmaceutical Sciences
(2013) from the University of Groningen. Her thesis was focused on the
effects of targeted delivery of interleukin-10 to the fibrotic liver. She also
has experience in parasitology and tropical diseases.
Edgar Phillips is a postdoctoral researcher in the Institut Jean Nicod at
the Ecole Normale Supérieure, Paris and an affiliate member of the
Thumos research group at the University of Geneva. He works in the phi-
losophy of mind, philosophy of action, and ethics. He gained his PhD in
Philosophy at University College London in 2018.
xii NOTES ON CONTRIBUTORS

Monica Roland holds a PhD in Philosophy from the University of Oslo,

where she was an affiliate at the Centre for the Study of Mind in Nature
(CSMN). Her main research interests are the philosophy of love (and
more generally the philosophy of emotions), normativity and the
nature of reasons, rationality (and its relation to irrationality), phi-
losophy of science, and feminism. In her dissertation What Is Love?
(2016) she defends the claim that love is a moral emotion, as well as
a response to reasons that justify the emotion. She currently works at
Oslo Metropolitan University (OsloMet) as a senior advisor.
Andrew Sneddon is a professor in the Department of Philosophy at the
University of Ottawa. He studies ethics and philosophical psychology. His
books include Autonomy (2013) and Like-Minded: Externalism and Moral
Psychology (2011).
Lotte Spreeuwenberg is researching love, ethics, and moral psychology
at the University of Antwerp in Belgium. She is writing her doctoral dis-
sertation Against the Fat Relentless Ego: Love at the Centre of Morality,
with Iris Murdoch as an overarching author. Besides publishing on love in
academic journals, she spends a great deal of her time writing public phi-
losophy, with a focus on activism. With Martha Claeys, she produces and
hosts the philosophy podcast Kluwen, which aims at showing that philoso-
phy deserves life beyond academia.
Ryan Stringer is a crazy cat-lover who earned his PhD in Philosophy
from the University of California San Diego in 2019 after completing and
defending his dissertation, The Nature and Normativity of Love and
Friendship. Besides the philosophy of love and friendship, he specializes in
metaethics, normative ethics, and moral philosophy in general. He is cur-
rently teaching at Purdue University for the Cornerstone Program.
Ryan Wittingslow is an assistant professor at the University of Groningen,
and holds a PhD in art history and philosophy from the University of
Sydney (2014). Most of his research sits at the intersection of aesthetics,
philosophy of technology, and philosophy of design. He also has devastat-
ing opinions about art.
 CHAPTER 1

Introduction

Simon Cushing

1.1   I Want to Know What Love Is

Love has been a topic of interest to philosophy since at least the time of
Plato’s Symposium, but with a few notable exceptions, it was unduly
neglected in the twentieth century, at least by writers in the analytic tradi-
tion that predominates in the English-speaking world. However, in the
past quarter century, writing on the topic has exploded. In this volume, we
touch on most of the currently hot debates and also introduce some fasci-
nating tangents. The main threads of discussion reflected in this volume
are as follows: the relationship between love and morality (is it adversarial,
congenial, or are they in fact co-dependent?); whether love is rational,
subject to reasons for or against it, or a force that is not under our inten-
tional control; and whether love affects the way we perceive the world or
the way we value things in the world (and whether this is a good thing).
More singular topics include: whether love would be affected by disputes
in the literature on free will; whether we could be mistaken about being in
love; whether our pets are capable of loving us back; whether a relation-
ship of the kind shown in the movie her between a human and an artificial

S. Cushing (*)
University of Michigan–Flint, Flint, MI, USA
e-mail: simoncu@umich.edu

© The Author(s), under exclusive license to Springer Nature 1

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_1
2 S. CUSHING

intelligence could be either loving or ethical; and whether the difference

between patriotism and nationalism hinges on how each instances a differ-
ent kind of love (and what that says about each of those “isms”). Along
the way we will see analyses of the work of philosophical greats like
Immanuel Kant as well as the work of more contemporary writers, in par-
ticular Iris Murdoch, and philosophers actively engaged in the current
revival, notably Harry Frankfurt, J. David Velleman, and Niko Kolodny.

1.2   (If Loving You Is Wrong) I Don’t Want

to Be Right

Love and morality may seem to be independent of each other, and often
even at odds. Othello, having murdered his beloved Desdemona out of
jealousy, says that he is “one that lov’d not wisely but too well,” implying
that love is a force that can propel one to commit monstrous acts. For a
more recent fictional example, think of Jaime Lannister pushing Bran
Stark out of a window while muttering, “The things I do for love.” The
fact that morality requires us to be impartial while love is very much partial
is at the core of the apparent tension. The influential post-war British phi-
losopher Bernard Williams is responsible for probably the most cited dis-
cussion of an illustration of the potential clash. He quotes Charles Fried’s
discussion of a man confronted with the choice of only being able to save
one of two drowning people, one of whom is the man’s own wife. Fried
argues that the man can be morally justified in saving his wife over a
stranger, but Williams bemoans even the need to give moral justification
for his partiality.

[T]he idea that moral principle can legitimate his preference, yielding the
conclusion that in situations of this kind it is at least all right (morally per-
missible) to save one’s wife… provides the agent with one thought too
many: it might have been hoped by some (for instance, by his wife) that his
motivating thought, fully spelled out, would be the thought that it was his
wife…. [T]he point is that somewhere (and if not in this case, where?) one
reaches the necessity that such things as deep attachments to other persons
will express themselves in the world in ways which cannot at the same time
embody the impartial view, and that they also run the risk of offending
against it. (Williams 1981: 18)
 1 INTRODUCTION 3

As we shall see, this case and Williams’ phrase “one thought too many”
have proved to be quite a touchstone in contemporary discussion. Williams
provides the basis for arguing that love itself provides reasons that not only
do motivate us independently of morality, but should do so, even in cases
of apparent conflict. This is a theme we shall see revisited in several papers
in this collection. However, many philosophers down the ages have argued
that, to the contrary, love and morality are intertwined, that you cannot
have one without the other. So we have two camps: one promoting love
as a force independent of morality and in some senses deserving to win out
over it in the battle to motivate action; the other arguing that they march
in sync and should not be seen as at odds. Confusingly, both camps can
cite the work of Immanuel Kant, usually held up as the greatest of the
modern philosophers, and certainly among the most influential ethicists,
for support for their position. On the one hand, Kant’s ethics are notori-
ously demanding: he argues for a system of exceptionless rules and con-
tends that one’s action only has moral worth if one acts out of duty,
prompted by one’s rational nature. This does not seem like a conception
of morality that would have room for love as we typically conceive of it: for
Kant, if one were to help a loved one simply because one loved them, this
would not count as a moral act. However, two of contemporary philoso-
phy’s most influential defenders of “love as a moral emotion” are explicitly
influenced by Kant. The first of these is Iris Murdoch, who was a philoso-
pher before she became known for her novels (and for being the subject of
the 2001 film Iris), whose work is the main subject of two chapters in this
collection. The other, whose work is cited in just about every chapter in
this volume, is J. David Velleman. He writes:

We have made a mistake… as soon as we accept the assumption of a conflict

in spirit. Love is a moral emotion precisely in the sense that its spirit is closely
akin to that of morality. (Velleman 1999: 341)

Because Velleman’s work has been so influential on the philosophy of

love of the past twenty-odd years, I will take a moment to outline his
view. In his initial essay, Velleman took particular issue with what he saw
as an emerging orthodoxy that love was a drive. Much of love’s bad
moral reputation, he argued, could be traced to Freud, who presented
love as a drive that all too often was based on misperception of its object.
Contemporary analytic philosophers followed Freud in presenting love
as a “syndrome of motives” with an aim, primarily desires directed
4 S. CUSHING

toward the beloved, typically to be with, to benefit, to please, or to be

well thought of by them. Instead of this, Velleman conceives of love as a
way of perceiving a loved one:

A sense of wonder at the vividly perceived reality of another person is, in my

view, the essence of love. (Velleman 2008: 199)

Velleman argues that love exactly parallels Kantian respect: for Kant, the
ultimate directive of ethics (the Categorical Imperative) insists that we
respect the personhood of others such that they must be regarded as
sources of value and never used as a means to one’s own ends. Kant argues
that respect for another acts as a check on one’s own tendencies to want
to exploit things in the world around one. Where respect is the mandated
minimum attitude toward other persons, love, claims Velleman, is an
optional maximum attitude, and love arrests our tendencies toward emo-
tional self-protection, leaving us vulnerable to the objects of our love. This
vulnerability can lead us to appreciate our beloved’s features so that we
may say that we love their crooked smile, but this does not mean that we
love them for their crooked smile (which creates all kinds of problems for
other theorists who do claim this, as we shall see), but rather we love the
smile because it is “an expression or symbol or reminder” of the beloved’s
value as a person. Velleman also stresses that while love is an attitude of
valuing another, it is not one that compares them to others. To love com-
paratively is to put a price on what we value, such that it can be replaced
by something of equal or greater value. But love does not rank beloveds
any more than parents rank their children. Again drawing on Kant,
Velleman argues that the kind of value that persons have is incomparable
because it responds to their dignity, not to a price. This, he says, is the
solution to the paradox that anyone equally may be loved, but the love for
each is uniquely special. Love is a moral emotion, just as respect is, because
it is a response to the dignity of persons. Evidence of its moral influence is
that it is the means by which moral lessons are first taught to children and
by which the moral sensibilities of adults can be (re)enlivened (Velleman
2008: 201). But perhaps most importantly, love is the emotion that makes
us care about the good of another, makes us work to ensure their flourish-
ing. Thus it is that we may end up desiring to help them, not because love
itself is that desire, but because wanting to help them is a natural result of
the vulnerability to them that itself comes from “really seeing” them (to
use a phrase Velleman borrows from Murdoch).
 1 INTRODUCTION 5

Another philosopher who offers a Kantian take on a phenomenon one

might have thought outside of the moral sphere is Neera Kapur Badhwar,
who argues that the only way to make sense of the fact that true friendship
(and love) must be constant through changes in personality and feeling is
to locate that friendship not in a response to any contingent, inconstant
features of one’s friend or lover, but in their humanity (Badhwar 1989).
Velleman’s conception of love and Badhwar’s of friendship are overtly
influenced by Kant, but are they pictures that Kant himself would recog-
nize or endorse? In Chap. 2, “Making Room for Love in Kantian Ethics,”
Ernesto Garcia explores this question.
First, Garcia points out that Kant actually discusses love in writings
other than the ethical works that influence the two contemporary think-
ers. Kant makes a distinction between moral and non-moral (or natural)
love, which he calls love as a passion. While both dispose us to help others,
they can be distinguished both by what aspects of our nature motivate
them (rational or “sensible”) and whether or not each can be morally
required of us (the former can, because it is a moral obligation, while the
latter cannot, because it is a contingent feeling, and thus not under our
voluntary control). If moral love sounds strange, Kant argues that in fact
it is the only way to understand the passages of the New Testament that
command us to love our neighbors and enemies, for non-moral love can-
not be commanded. Kant also makes a parallel distinction between types
of friendship, except that moral friendship also incorporates moral love.
Now, with this distinction in Kant, it appears that he has acknowledged
that there is a legitimate form of love that is non-moral, and that perhaps
his moral love is a specialized kind, a religious ideal. Has Velleman, in
arguing that love of the kind that we might feel for our children and our
partners is fundamentally a moral attitude, conflated the two loves?
A problem for Kantian moral love is that it must be truly universal: we
are required, after all, to love even our enemies. The love that Velleman
and Badhwar want to defend, however, is the actual, partial (in the sense
of “not impartial,” rather than “incomplete”) love we feel for a select few
individuals. How can Velleman and Badhwar explain this partiality on our
parts? Badhwar argues that it is not the humanity that is found within
every person that our friendship responds to but rather the empirical per-
sonhood of concrete historical individuals. Velleman argues that the love
he is defending is optional (unlike the minimum required attitude of
respect), and its reasons can be particular. But what it is is a disarming of
emotional defenses in response to the incomparable and non-instrumental
6 S. CUSHING

value of another human being. Garcia concludes that, on the one hand,
the modern Kantians’ accounts of love are improvements on Kant’s own
“moral love,” because we find the notion of loving someone out of duty
counterintuitive, and the modern Kantians place the focus of love in the
right place, that is, on the people who are the objects of our love and
friendship. However, Garcia argues that what Kant gets right is in separat-
ing love out into kinds, at least one of which is non-moral. Tackling
Velleman’s account in particular, Garcia argues that it is open to a strong
and a weak reading. The weak reading is trivially true (love involves mak-
ing oneself emotionally vulnerable to another who is a being worthy of
moral respect) but does not show that love is moral. The strong reading,
on the other hand, which requires that we love someone because of our
knowledge of them as a moral agent, is neither sufficient for love (it is
equally true of appealing for help, mercy, or friendship) nor necessary (one
can love another romantically without viewing them as a moral agent).
Garcia concludes by agreeing with Berit Brogaard that there is not a single
kind of love, and while Velleman-style moral love might indeed be one
kind of love, there are others, and others of value.

1.3   The Look of Love

Before Velleman, and in fact influencing his argument for love as a moral
emotion, was the British philosopher-turned-novelist, Iris Murdoch,
whose work is undergoing something of a revival. Cathy Mason’s chapter
presents both an argument in support of her account and a critical evalua-
tion of writers who, while inspired by Murdoch, have abandoned what
Mason takes to be core parts of Murdoch’s view. A primary motivation for
Murdoch’s writing was that the dominant behaviorist approach of mid-­
twentieth-­century philosophy ignored a vital kind of moral activity that
was purely internal. Prior to any action that can be deemed of moral
worth, we must, argued Murdoch, practice attention, a fundamentally
moral attitude to the world that is a kind of love. This attitude has an epis-
temic aspect: only through viewing the world with love can one achieve a
true understanding of reality, and it is this process of attending to the
world in a loving way that is itself a moral activity. In a key example,
Murdoch describes how a mother-in-law (M) with an initial dislike of her
son’s wife (D) manages to revise her view after “looking again.” In so
doing, not only does she arrive at a truer picture of reality, she herself is
improved by the process. While contemporary writers defending love as a
 1 INTRODUCTION 7

moral attitude have been influenced by Murdoch, Mason argues that they
do not do full justice to important elements of her view. For example,
Velleman’s view of love as an appreciation of the moral personhood of
another that requires “really seeing” one’s beloved departs from Murdoch’s
approach both because Murdoch, unlike Velleman, insists that love is mor-
ally necessary, and not merely optional, and because, for her, it is the con-
crete particularity of an individual that love focuses on, not the rational
will that every person instantiates equally, as in Velleman’s account.
Mason also considers Mark Hopwood’s sympathetic exegetical work on
Murdoch (Hopwood 2014, 2017) and finds that while it acknowledges
Murdoch’s view of the particularity of the subject of love and describes an
epistemic role for love, that role is not the one Murdoch intended.
Hopwood says love reveals normative demands on the lover, but, Mason
contends, Murdoch insists that love’s role is primarily to reveal facts about
the person being loved. Furthermore, the facts revealed by the loving gaze
are both objectively real and unable to be captured in the supposedly
value-neutral language of science. Murdoch, argues Mason, views love as
a character trait, as, in fact, a virtue, alongside those studied by the ancient
Greeks, including courage and wisdom, and like those character features,
love is a reliable sensitivity to real features of the world. When one gazes
on another with love, as M did with D, the good qualities that one’s
beloved genuinely possesses (and not qualities that one projects on them
because of one’s loving gaze, as an anti-realist might contend) are revealed.
As might be unsurprising, Murdoch’s view of love has appeared quix-
otic to some critics. Does it really map on to love as we normally under-
stand it? Against the criticism that Murdoch’s epistemic conception of
love rules out the affective component that is stressed in all love songs, for
example, Mason points out, first, that there are respectable theories of
emotion that present all emotions as having an epistemic component; sec-
ond, that our common-sense conception of love includes the thesis that
true love requires truly knowing one’s beloved; and third, that love cannot
be reduced to an affective state alone, because such states are necessarily
intermittent, whereas a love can last a lifetime. Against the criticism that
Murdoch’s view cannot account for the selectivity of love—a criticism lev-
eled at Velleman’s view, as we saw, and potentially worse for Murdoch,
who rejects the idea that love is optional—Mason distinguishes between
love and loving attention. Whereas the latter might be what is com-
manded, it is necessary but not sufficient for the variety of loves that there
are, and what might make particular loves (for one’s children, for a
8 S. CUSHING

romantic partner) selective or unique might be a function of how one’s

relationship or the behavior of one’s beloved facilitates the loving atten-
tion. Finally, there is the criticism that Murdoch’s view seems to presume
that everyone is a suitable subject for loving attention. But we are familiar
with loves that we think are profoundly mistaken, or individuals who make
unsuitable subjects for loving attention, perhaps because they are irre-
deemably wicked. But, first, a truly loving attention, because it is attuned
to reality, would reveal an absence of good just as much as its presence.
And second, the idea that loving attention is commanded for all is not
peculiar to Murdoch: its most famous proponent is the Jesus of the
Gospels.

1.4   Love Me Do
The philosopher Sally Haslanger, in her writing on gender, coined the
term “ameliorative inquiry” for an approach to defining a concept that
aims not solely at descriptive accuracy about the way people currently use
it, but at producing a possibly revisionist, improved version, with the aim
of making the society that employs the concept a better place. In Chap. 4,
“‘Love’ as a Practice: Looking at Real People,” Lotte Spreeuwenberg sug-
gests that we should do the same for the concept of love. As we have seen,
both Velleman and Iris Murdoch have offered influential moralized
accounts of love, and Spreeuwenberg’s first task is to evaluate whether
either suits the ameliorative inquiry she has in mind. Tackling Velleman’s
first: while she applauds his commitment to “really seeing” one’s beloved,
she, in common with many of Velleman’s critics, finds unsatisfactory both
his view of the subject of the loving gaze as the Kantian self, and his solu-
tion to the selectivity of love (in contrast to the universality of the require-
ment of respect) in the claim that it is a contingent fact that we respond
with love to certain empirical selves and not to others. The problem with
this, points out Spreeuwenberg, is that it fails to account for the personal
character of love, both because the bare Kantian self is impersonal, but
also because it is a mystery what features of an individual may trigger us to
respond to their Kantian self rather than another’s.
Spreeuwenberg thus turns to the view of an author who attempts to fix
this flaw while preserving what is valuable in a Velleman-like approach:
Pilar Lopez-Cantero. She suggests that the subject of the loving gaze is
not the bare Kantian self but its product, which is a narrative. This is
indeed unique to each person, thereby better accounting for the personal
 1 INTRODUCTION 9

character of love, with “narrative fit” between lover and beloved explain-
ing when and why love blossoms. However, Spreeuwenberg finds the
views of both Lopez-Cantero and Velleman to be too passive, certainly for
her ameliorative inquiry, and uses the example of Dante and his “muse”
Beatrice (whom he barely exchanged two words with, but fixated on) to
illustrate why. While Dante certainly believes he loves Beatrice, and is
caused to do so by some feature he perceives in her, he is not perceiving
her as she truly is, but as some ideal that he projects on to her. Love should
be, in the words of bell hooks, a verb, that is, active, a process, as Adrienne
Rich puts it, “of refining the truths [lovers] tell each other.” Viewing it
this way shifts the focus from the lover alone to an interactive partnership,
and Iris Murdoch’s writings on love provide a framework for this active
approach. Murdoch’s “M and D” case, described earlier, is an illustration
of love as truly attending to the target of one’s gaze to see her in her
(changing) reality. Spreeuwenberg considers suggestions by psychologists
like Lisa Bortolotti that projecting fantasies on to one’s partner might
have positive effects for the lover or the relationship, but concludes that
fantasies are no part of the ameliorative project, especially if we widen its
scope to the political sphere and call on love to break down barriers
between oppressors and oppressed. Spreeuwenberg ends by cautioning
that we should not assume that we can capture the full reality of our
beloved, or even that this is the goal, agreeing with Carla Bagnoli that
understanding another has a possibly invasive aspect. But love as attending
to others is the love that will make us and our society a better place, and
therefore the best reconstruction of the concept for a project of making
love a force for good.

1.5   Will You Still Love Me Tomorrow

It is a truth almost universally acknowledged that love is something over
which we have no control: we fall in love, sometimes against our better
judgment and to our own dismay. On this view, love is, in the words of the
title of a paper by Nick Zangwill, “gloriously amoral and arational”
(Zangwill 2013). However, I say “almost,” because in Chap. 5, “Love,
Choice, and Taking Responsibility,” Christopher Cowley argues for an
important role for choice in love, in the form of a lover taking responsibil-
ity for meeting the prospective needs of their beloved. Thus, Cowley joins
the previous authors in finding a moral core in love. Cowley takes up
Susan Wolf’s suggestion that there is a virtue in a willingness to take
10 S. CUSHING

responsibility for what one has yet to do (prospective responsibility, rather

than the more usually discussed retroactive kind) and locates just such a
virtue in the commitments of an ongoing loving relationship between
adults. Using the marriage vow as a case study, Cowley argues that it
involves not just responsibility for one’s spouse and the needs of the per-
son they will become, but also responsibility for becoming the kind of
person who will stay committed, both to the love and to the meeting of
needs. Cowley responds to challenges to this analysis of love both from
existentialists, who could argue that it involves an abandonment of radical
freedom and a bad faith essentialization of both parties, and also from a
famous case described by Michael Stocker, where a person confined to a
hospital is dismayed to find that the person he had considered a friend was
motivated solely by moral duty to visit him (Stocker 1976). Cowley sug-
gests that we can distinguish between duty and responsibility, where the
problem with the former is that it is impersonal, while the latter comprises
a response to the particular friend and their needs. In this respect, Cowley’s
view resembles Harry Frankfurt’s depiction of love as a configuration of
the will that presents the lover with felt necessities that they experience
just like the demands of conscience. However, Cowley finds Frankfurt’s
view too unilateral, as it focuses entirely on the experiences of the lover,
allowing for such phenomena as unrequited love or even love of non-­
persons. Cowley contends that the love he wants to defend is necessarily
instantiated in the relationship between two lovers, both of them moral
agents. That love, claims Cowley, is depicted beautifully in The Little
Prince, where those who have chosen each other become unique in the
whole world to each other. In some respects, Cowley’s view solves the
problem of specificity (that we saw plagued Velleman’s view, in the eyes of
his critics) in a similar way to Niko Kolodny’s view (Kolodny 2003), but
adds to it the normative principle that love requires of us that we maintain
and live up to the demands of that relationship.

1.6   Love Is a Battlefield

Just how seriously should we take the comparison between love and war
at work in both Ecclesiastes and the well-known proverb? Andrew Sneddon,
in Chap. 6, “Not All’s Fair in Love and War: Toward Just Love Theory,”
suggests that, just as “Just War” theory subverts that proverb in the case
of war, we should construct a “Just Love” theory to provide an ethical
roadmap for loving relationships. Just War theory is typically divided into
 1 INTRODUCTION 11

three parts: jus ad bellum, which concerns the grounds for going to war;
jus in bello, which lays out the restrictions on what is acceptable while war
is waged; and jus post bellum, covering the aftermath. Sneddon focuses on
analogs of the first two for the conducting of loving relationships. But
before embarking on either, he first contends with the challenge faced by
any attempt to lay out the ethics of war: that the very idea is naïve. Self-­
styled “realists” reject the notion that war is an appropriate (or even pos-
sible) subject of a code of ethics. A parallel challenge to Just Love theory
takes Zangwill’s stance on the amorality of love. However, even were that
the case, Sneddon contends that loving relationships, and the actions taken
in their context, are very much intentional, and thus subject to moral
evaluation. Furthermore, Sneddon argues that if we assume the following
things about love (he focusses strictly on the romantic kind): that it is
other-directed, tied up with other emotions, and love affects other emo-
tions holistically, so that emotions felt as part of a loving relationship are
experienced as part of that relationship, this reveals the need for an ethical
rulebook, given how profoundly one’s actions affect the other party in a
loving relationship.
The love analog of jus ad bellum Sneddon calls jus ad amantes necessi-
tudo. Where war requires a just cause, love requires a just target, someone
who is capable of participating in a loving relationship and capable of con-
senting to the costs of that relationship. The costs may depend on the
goals of the relationship, which must also be just. These goals can be
internal to the relationship, such as being partners, or external (in the
sense that they could possibly be secured without such a relationship), like
having sex, children, or company. Other criteria of jus ad amantes necessi-
tudo include “necessity” (a loving relationship is necessary to achieving
the goals, at least, when they are internal), “proportionality” (of the rela-
tionship to the strengths of one’s sentiments and importance of the
goals—interestingly Sneddon allows that if one is infatuated with a celeb-
rity, seeking a relationship with them is not ruled out by this criterion,
although very likely by others), and “chance of success” (the analog of
chance of victory in war). One tentative conclusion Sneddon draws is that
it will be very difficult for external goals to justify a relationship according
to these criteria—so much for arranged marriages.
What about rules for behavior within loving relationships? Jus in aman-
tes necessitudo governs actions motivated by love in a relationship already
established, and, argues Sneddon, must be weighed against other priori-
ties in a life well-lived. Displays of affection that bother others (one thinks
12 S. CUSHING

of the Seinfeld episode (“The Soup Nazi”) where Jerry and his girlfriend-­
of-­the-episode (played by Alexandra Wentworth) refer to each other as
“schmoopy”) are out, and Sneddon recalls bitterly having to cover for a
co-worker at a fast food job because she was trying to reconcile with her
boyfriend. How useful is this sketch? Can real lovers actually follow such
guidelines, or is the realist right to scoff? You be the judge, dear reader.

1.7   Is This Love?

It is impossible to doubt whether or not one is in pain. But it does not
seem to be impossible to doubt whether or not one is in love. In Chap. 7,
“Doubting Love,” inspired by Graham Greene’s novel The End of the
Affair, Larry A. Herzberg analyzes the nature of love that would make
this doubt possible, and what are the limits of doubt and certainty in mat-
ters of the heart. Herzberg draws inspiration from R.J. Sternberg’s influ-
ential “triangular theory” of love that divides love into distinct components
of emotion, passion, and commitment. He argues that doubt is possible to
varying degrees about each of these components. Least doubt is possible
about our passionate feelings (they are most similar to feelings like pain,
whose presence or absence is indubitable), most doubt is possible about
our emotions, and somewhere in between lies our certainty about the
nature or existence of our commitments. Herzberg agrees with Christopher
Cowley on the importance of commitment to love and points out that
there has to be a volitional element to love to explain both the defensive-
ness and guilt about the wrong answer to the question “do you love me?”
as well as feelings of betrayal against a lover whose love goes away.
Herzberg also argues that one may doubt one’s emotions, both because
one may be unsure of their objects, and because love can be comprised of
a cluster of other emotions, each of which may be hard to distinguish from
other, closely related emotions which are not indicative of love. However,
Herzberg concludes that doubt is not an essential corollary of love and
that there are many people whose circumstances and history ensure that
they can be sure of their love.

1.8   Can’t Help Falling in Love

The view that love is something that happens to us irrespective of our
plans and choices has fared poorly so far, despite its intuitive appeal, and in
Chap. 8, “Love and Free Agency,” Ishtiyaque Haji aims to deliver another
 1 INTRODUCTION 13

wounding blow against it. Haji argues that love is “fragile,” in the sense
that the value or even existence of love is conditional on the results of age-­
old philosophical debates about free will. To put this in context, since at
least the time of the ancient Greeks, arguments have been considered that
purport to threaten our usual conviction that we are free agents, able to
control our own destinies, and, as a corollary, appropriately subject to
assessments of responsibility such as praise and blame. That is, the reason
why we standardly think it is legitimate to hold people accountable for
their wrongdoings (or praiseworthy for their virtuous acts) is because we
think those actions were up to them, under their control, not merely things
that happened to them. However, “responsibility skeptics” produce argu-
ments to show that, really, our actions are not up to us, and our belief that
they are is based on an illusion.
Extreme responsibility skeptics argue that none of our actions are ever
up to us, which seems very radical to the uninitiated, but one can work
gradually toward that conclusion by less extreme steps. One such step is to
argue for “responsibility historicism,” which is the view that whether or
not somebody is capable of the kinds of action that merit assessments of
responsibility depends on factors outside of the mind and body of that
individual (which is why the view is also called “externalism”). One major
argument for externalism involves thought experiments depicting fiendish
psychological manipulation of individuals, such as Alfred Mele’s example
“One Bad Day” (Mele 2019: 20–21) quoted by Haji. In this case, the
saintly Sally is manipulated to have just the same evil psychological makeup
as the merciless murderer Chuck, so that Sally intentionally plots and exe-
cutes a hapless victim over the course of the titular day, only to have her
saintly psychology reinstated by the same twisted psychological manipula-
tors during the night that follows. Mele contends that we should all agree
that Sally should not be held responsible for her murder, but Chuck should
for his, even though both are the same from an internalist perspective.
Thus, whether or not one should be held responsible for an action depends
on the history of how the psychology that produced that action was
acquired (hence “historicism”). Canvassing various recent philosophical
accounts of love (some of which should be very familiar to us by now),
Haji contends that all of them contain necessary psychological elements
that are open to the same kind of arguments for externalism as the
responsibility-­undergirding ones in “One Bad Day.” To make the point,
Haji describes “One Lovely Day,” where whatever psychological states
manifest Romeo’s love for Juliet are implanted in Romello for a day.
14 S. CUSHING

During that day, asks Haji, “does Romello indeed love Juliet?” Haji con-
tends that if we are moved by the externalist arguments supported by
“One Bad Day,” then we should conclude that Romello does not, and
that love, like responsibility, depends not just on the presence of certain
psychological states, but on how they have been acquired. (Lest one won-
der about the relevance of paranoid science-fiction cases involving devious
covert mind-manipulators, once historicism is established, the next step is
to argue that natural forces like genetics and environment, the kinds of
things that really do shape our psychological makeups, can have similar
responsibility/love undermining effects.)
However, even if one is not convinced by this to become an externalist,
Haji maintains that even from an internalist/anti-historicist standpoint,
“One Lovely Day” reveals three results. First, even if Romello really does
love Juliet during his manipulated day, that love is of a lesser value—is
forced or ersatz. Second, to be an instance of loving behavior, an action or
state has to issue from love. This parallels the distinction, insisted on by
Kant, among others, between praiseworthy virtuous action from duty, and
non-praiseworthy, only apparently virtuous action in accordance with duty.
That is, just as one’s helping somebody is not meritorious if one does it
solely for an expected reward, so one’s showing affection to another is
only praiseworthy from love’s standpoint if it is motivated by love, and not
by duty or other considerations. Haji develops this thought in a section
where he expands the suggested analogs between behavior that is morally
responsible and that is motivated by love, arguing that any view of love
that posits that “emotions may be construed as constituting relationships
of love and friendship” supports this parallel. Haji goes on to propose the
notion of normative standards (he suggests the terms “commendability”
and “censurability”) from love’s standpoint. The third result Haji adduces
from “One Lovely Day” is that love is fragile in the sense introduced at the
start, that it has “freedom or autonomy presuppositions.” Finally, Haji
considers an attempt by noted free-will skeptic Derk Pereboom to save
love from just the kind of externalist considerations that he (and Haji) use
against responsibility. Pereboom is, in effect, a “love optimist,” because he
believes that while there may be emotions, like remorse and guilt, that are
both associated with relationships and “fragile” to externalist consider-
ations, they may easily be substituted by non-fragile alternatives, like sor-
row and regret. Against this attempt, Haji contends that these suggested
analogs are equally fragile. If Seth harms you, suggests Haji, and then
expresses sorrow, you would not accept that sorrow as genuine if you
 1 INTRODUCTION 15

found that it was (once again) produced in him solely by psychological

manipulation. Thus Haji concludes that whether or not we are free agents
impacts not just whether or not we are morally responsible, but also
whether or not we are capable of love, something philosophers of religion
have long insisted on in offering justifications for God granting us free
will, despite the fact that (they claim) it is responsible for all the evil and
suffering in the world.

1.9   Love Me for a Reason

A theme running through the chapters we have just discussed has been the
dispute over the extent to which love can be seen as a “gloriously ara-
tional” force that overtakes us and carries us along in a way that is beyond
our control (and is thus something we cannot be held accountable for) or
whether or not our love and what issues from it is a matter of rational
appraisal. The front on which this dispute is most overtly fought in con-
temporary analytic philosophy is in the dispute over the relationship (or
not) between love and reasons.
The most influential philosopher advocating a “No Reason” view of
love (which, to be clear, means that no reasons can be given for why we
love, although, as we shall see, he argues that love itself is the ground of
reasons for a great many things) is Harry Frankfurt. His writing provided
the motivation for many critics who have themselves become influential,
most notably Velleman and Niko Kolodny. Such critics point out that love
does not seem like bodily functions such as sweating and digestion, things
that are genuinely not the products of our decisions. And it does seem
like, if asked “why do you love x?” the kind of answer expected is not sim-
ply “because certain chemicals were released in my brain” but rather one
involving reasons like “x is so dreamy!” or “x is my child!” or even, if one
believes Velleman, “x is a Kantian moral agent.” The love one feels, in
short, can be justified. Or so goes the Reason View of love. There are sev-
eral variants, of course, but the most intuitive has it that when x loves y it
is because of a feature or set of features of y. On the one hand, this seems
right: x’s love of y should be explained by something about y. If the reason
for x loving y had nothing to do with y, then y would no doubt be insulted
(“I love you because of a compulsion I have to love everybody”). And, of
course, we saw that a challenge faced by Velleman’s view was that he
picked a feature that was too general (at least for his critics—as Kolodny
put it, “personal ads do not read: ‘Bare Kantian person seeks same’”
16 S. CUSHING

(Kolodny 2003: 174)), but on the other hand, if one does pick particular
features of the beloved, then that also seems objectionable. Philosophers
writing on this topic are fond of citing W.B. Yeats’ “For Ann Gregory,”
wherein a girl with gorgeous “honey-colored” hair yearns that young men
“May love me for myself alone/And not my yellow hair.” This is the core
of the No Reason view’s attack on the Reason view, but No Reasoners
have more weapons in their arsenal. In Chap. 9, “Sentimental Reasons,”
Edgar Phillips, citing Setiya (2014), lists four puzzles that point to appar-
ently counterintuitive implications of a Reason view. First, universality: if
Ennis’ love for Jack is based on good reasons, shouldn’t every rational
agent, exposed to the same reasons, also love Jack? Second, promiscuity: if
Catherine loves Jules for a certain reason (say, the insouciant way he
smokes his Gauloises), then if Jim embodies the same feature, shouldn’t
Catherine also love him? Third, trading up: suppose Kamariah loves
Thomas for his long curly hair. If someone with an even more impressive
mane shows up, it implies that she should abandon Thomas for the prefer-
ably coiffed alternative. Finally, inconstancy: philosophers who discuss this
problem are wont to cite Shakespeare’s Sonnet 116 wherein he writes
“Love is not love/Which alters when it alteration finds.” When Billy Bragg
says, “And then one day it happened/She cut her hair and I stopped lov-
ing her” (“Walk Away Renee (Version)”), it is not meant to reflect well on
the maturity of the authorial voice, but it seems to follow from the Reason
View that should you lose the features that were the basis for my love for
you, then my love will cease.
Our two authors on this topic, Phillips and N.L. Engel-Hawbecker,
respond to puzzles such as these by digging deeper into the nature of rea-
sons themselves. Phillips points out that there are actually three different
kinds of roles that reasons can play. Reasons can explain behavior by point-
ing to a cause, whether or not that cause was known to the actor. Reasons
can also be what a person has in mind when acting intentionally (“moti-
vating” or “personal” reasons). Finally, reasons can justify one’s behavior.
It may be, posits Phillips, that a particular Reasons View of love envisages
the kind of reason in question to play one of the roles, but not the other.
For example, the fact that I was hungry is a perfectly good explanation of
why I bought a loaf of bread. Raising the objection “but that doesn’t
explain why you bought that loaf of bread!” seems beside the point.
However, he concludes that for many of the proponents of Reason Views,
the reasons are meant to play more than one of these roles. So next he
suggests that perhaps the problem is that we are mischaracterizing love by
 1 INTRODUCTION 17

comparing it with reason-responsive dispositions like intention or belief.

While it is true that a belief I have (say, that injecting bleach will cure
Covid-19) requires justification, and should alter if the circumstances it
concerns change or new information comes to light, this does not neces-
sarily apply to love. Perhaps, suggests Phillips, love is a sentiment, where
such things, like one’s character, are deep, long-lasting, developed gradu-
ally over time, and not formed by choice. This is not to say, however, that
Phillips is defending a pure No Reasons view: reasons are too important to
interpersonal affairs to be abandoned entirely, and once again, love is not
like digestion. However, the kinds of reason one gives may not be profit-
ably judged by the kinds of reason one should give to justify one’s beliefs
or intentions.
In Chap. 10, “Wouldn’t It Be Nice? Enticing Reasons to Love,” Engel-­
Hawbecker takes a different tack. He frames the No Reasons view’s chal-
lenge as the assertion that the following two claims cannot be held
simultaneously: Requiring Reasons (reasons can require people to do what
they favor) and Love’s Prerogative (there is nothing that can require us to
love anyone). It is because of the former that we are presented with puz-
zles like promiscuity and trading up: if my love for x is explained by reason
r, and reason r applies also, or more so, to y, then I am required to love y
as well or instead. And it is because of the latter that we reject this conclu-
sion. The problem is not with Love’s Prerogative, but perhaps we can
reject Requiring Reasons without rejecting the idea that love is for a rea-
son. One attempt at this approach suggests that the reasons for love are
“warranting” reasons: reasons that, while they permit an option for which
they provide a reason, as do normal reasons, do not forbid its absence. If
this were the case then, for example, we avoid promiscuity because, while
one is permitted to love every bearer of the feature that is the reason for
love in one case, one is not required consistently to love each one.
However, Engel-Hawbecker finds two problems with this thesis. The
first is that, lacking an argument that all reasons for love are like this, it
appears ad hoc. But second, if it is the case that love needs reasons before
it is permitted, then something must be forbidding it otherwise. But one
can hope to be loved even if one admits that one lacks lovable features
without this hope being perverse, implying that there is nothing that for-
bids love even without features that might serve as reasons for it. So if
there are reasons for love that are not requiring, they are not warranting
reasons, says Engel-Hawbecker. Instead, we should notice that reasons
typically have two kinds of properties. When they require or forbid
18 S. CUSHING

something (as my belief that a number is odd requires my belief that it

cannot be divided into two equal integers), they are showing their deontic
properties. But they also have evaluative properties, as when they make
“an option attractive rather than demanded, required, or right,” in the
words of Jonathan Dancy (2004: 91), who first asserted the possible exis-
tence of reasons which have only evaluative properties, which he dubbed
“enticing reasons.” Engel-Hawbecker suggests that the reasons for love
could be just such kinds of reasons, and if so, that means we can abandon
Requiring Reasons and keep Love’s Prerogative, without being forced to
adopt a No Reasons view. Puzzles like promiscuity can be avoided: while
my love for x is justified by a particular set of properties they instantiated,
I am not thereby required also to love y because they also instantiate those
properties. And indeed, while reasons for love are, in his view, enticing,
Engel-­Hawbecker insists that that does not mean that reasons of love,
which may include duties to our current lover (x), cannot be requiring,
and may preempt being drawn away by a similarly featured other.

1.10   Drowning in the Sea of Love

In chapter 11, “Love, Motivation, and Reasons: The Case of the Drowning
Wife,” Monica Roland takes up the discussion of such reasons of love, the
kind of reasons that Harry Frankfurt does endorse. I have already had cause
to mention Bernard Williams’ discussion of Charles Fried’s “case of the
drowning wife” (and his contention that Fried gives the husband “one
thought too many” in deliberating about preferring his wife over another
who is also drowning), and this case is much-discussed in the contempo-
rary literature on love. Roland uses this case as a touchstone to discuss the
nature of the reasons love offers for benevolent acts toward others and
their relationship with moral reasons. She argues that Frankfurt, Velleman,
and Kolodny each get something right about the case and something
wrong. What Frankfurt gets right is that it is not the bare spousal relation-
ship that provides reasons for partial treatment, and further that the hus-
band does not need to reflect in the heat of the moment to be motivated
to help his beloved. What Frankfurt gets wrong, in Roland’s view, is that
loving relationships can be normatively significant, and, furthermore, valu-
ation of those relationships by the lovers can (partially) constitute the love
they have for each other. What Velleman gets right (and Frankfurt denies—
although Roland argues that he is undermined in this denial by his insis-
tence that love is a disinterested concern) is that love is a moral emotion,
 1 INTRODUCTION 19

necessarily involving the moral attitude of respect, and furthermore that

the reason for the husband’s partiality is their relationship. But Velleman
is wrong to say that the relationship plays no part in their love.
Both Frankfurt’s and Velleman’s accounts fail to provide an adequate
account of love’s selectivity—why one has reason to love only one’s lover
and not a qualitatively identical doppelganger. Kolodny’s solution to this
problem is that one is only in a relationship with the original, not their
clone: it is not the intrinsic properties of one’s beloved that explain the
selectivity of love, but the relational ones. Roland also agrees with
Kolodny’s insistence that love is deemed by the lover to be rendered
appropriate by the relationship (loving behavior by a stranger is inappro-
priate and disconcerting). But she believes that he is wrong to omit lovers’
mutual appreciation of both intrinsic and relational properties as partly
constitutive of the relationship. Roland ends up endorsing a “dual
account” of love. Velleman’s moralized valuation of personhood provides
one element, but is too general to suffice alone and must be comple-
mented by a valuation of particulars, including the relational properties
one’s beloved bears toward oneself. This results in the husband having not
one but two reasons to rescue his wife. But, to avoid a Williamsesque
charge that this gives the husband two thoughts too many, Roland sug-
gests that neither need be consciously formulated at the time of action:
“awareness of the inherent value of his wife and the special relationship he
has with her are built in to the very fabric of the husband’s dispositions
and thus implicit to his motivating thought.”

1.11   The Love Cats

Up to this point we have only considered love as something that happens
between (two) human individuals. However, many of us genuinely feel
that we love our pets, while at the same time acknowledging that they lack
many of the capacities that feature in several of the theories we have can-
vassed so far. (I have no doubt my cats do not respect me in the way
Velleman means, and possibly in any way.) In Chap. 12, “Can Our Beloved
Pets Love Us Back?” Ryan Stringer investigates whether there is any pos-
sibility that the behavior we take as affection in the non-human animals
(NHAs) in our lives could be indicative of something deserving to be
called love. It turns out that there are a number of books and articles writ-
ten by scientists that defend the claim that, indeed, some NHAs are capa-
ble of love, and do love the humans in their lives. While Stringer professes
20 S. CUSHING

himself keen to have this be true, particularly of the cats with whom he is
in an otherwise mildly abusive (on their part) relationship, as a philoso-
pher he feels he cannot take the scientists’ purported evidence as sufficient
without challenge. Against Gregory Burns’ (2013) claim that dogs’ empa-
thy for us is sufficient to demonstrate their love for us, Stringer points out
that one can feel empathy for someone whom one hates, and it might even
help in the task of making them suffer. Against Carl Safina’s claim (Dreifus
2019) that dogs’ desire to be near us for no other reason than to be near
us evinces their love, Stringer points out both that stalkers can have this,
and that it is in theory possible to have that desire isolated from any love
for the target of that desire. Finally, Stringer assesses several different pur-
ported pieces of evidence for canine love in Clive Wynne’s (2019) Dog Is
Love: Why and How Your Dog Loves You. That dogs have the capacity to
form affectionate relationships with us does not suffice, because such
things come in a wide spectrum, only some of which are loving relation-
ships. That dogs exhibit hyper-social behavior fails because there are con-
ditions that some humans have, including Williams-Beuren Syndrome,
which are similar but not taken as proof of love. Wynne fares better in
Stringer’s estimation by stressing that dogs show distress at being sepa-
rated from their humans, find it rewarding to be near them, and appar-
ently care about them to the point of trying to help them when in distress.
Of these, evidence of attachment is deemed too self-interested to count,
but Stringer takes the caring as the best potential ground for an attribu-
tion of a capacity to love.
So what does love consist in, if not these scientists’ criteria? Stringer
postulates that, whatever else comprises love, it must have at least the fol-
lowing three essential components: a disposition to feel affection (which is
more than the simple presence of affection, because love is more persistent
than such a potentially fleeting and necessarily intermittent feeling), a
non-instrumental concern for the welfare of one’s beloved, to the extent
of prioritizing the promotion of their welfare, and the assessment of one’s
beloved as so special as to be irreplaceable. Failure to capture all three of
these key components dooms the initially promising philosophical
accounts of Thomas Hurka (an attitudinal-dispositional theory) and of
Andrew Franklin-Hall and Agnieszka Jaworska (solely dispositional), but
these are potentially captured by Sam Shpall’s (2018) tripartite theory of
love. Shpall analyzes meaningful love as a devotion to an object that is
liked, which partly consists in special concern for that object’s good, which
partly consists in emotional vulnerability to that good and what affects it.
 1 INTRODUCTION 21

Stringer suggests that if the notion of devotion is expanded to include the

idea that the object of one’s devotion is irreplaceable, then this view cap-
tures his requirements. However, does it allow that NHAs can love the
humans that love them? Stringer concludes that if we allow that some-
thing that does not quite rise to Shpall’s standard of meaningful love is
still love, then dogs are plausible candidates for a capacity to love. He is
forced to conclude, however, that cats fail to meet the standards of emo-
tional vulnerability to our welfare and benevolent desire for our happiness.
In a coda, however, he suggests that the relationship we can have with cats
is still valuable and love-like, and this is no small thing.

1.12   Computer Love

So much for animals loving us—what about things that are apparently
inanimate? The 2013 Spike Jonze film her depicts a love story between a
human and an artificially intelligent operating system. Assuming such a
thing were possible, would it be desirable? Would the “love” that resulted
be of any value? In Chap. 13, “Romantic Love Between Humans and AIs:
A Feminist Ethical Critique,” Andrea Klonschinski and Michael Kühler
argue that there are important reasons to doubt that it would. The first
problem with the kind of relationship depicted in the film is that the “rela-
tionship” reinforces pernicious gender stereotypes. “Samantha” is created
to meet every need that Theodore might have. However, lest it be said
that this can be avoided by making AIs male, besides the fact that the male
ones can also reinforce stereotypes (particularly if they’re built into things
like your GPS and thus telling you what to do), the second problem rears
its head: because AIs lack autonomy, the relationship is of necessity asym-
metrical. This is true no matter what philosophical model of love you
favor; the authors consider models whereby love is construed as an atti-
tude instantiated in the lover, as in Harry Frankfurt’s love-as-caring model,
love is construed intersubjectively, as in Angelika Krebs’ model of love as
interpersonal sharing, or love is construed as a union, where the partici-
pants form a “we-identity,” as postulated by Mark Fisher, and Roberts
Nozick, and Solomon. In no case can the relationship between humans
and AIs meet the requirement of a love between equals, and thus, if it
counts as love at all, it is only a degenerate form, not worth pursuing.
Moreover, these problems cannot be solved simply by better program-
ming: giving the AIs a personality would not solve the power imbalance.
The user would still be able to adjust that personality in the “settings,” the
22 S. CUSHING

better to suit their preferences. And giving the AI actual moral autonomy
is either impossible (depending on your metaphysics) or potentially cata-
strophic. As Klonschinski and Kühler wryly note, it would not be finan-
cially advantageous to make a product that could reject its user, not to
mention the Terminator/Robopocalypse/Ex Machina apocalyptic possibili-
ties. Finally, lest the problem of unequal relationships with AIs be dis-
missed as ethically trivial, given that they are not persons, Klonschinski and
Kühler remind us of the deleterious effects on our relationships with per-
sons, and on our moral characters, particularly if the gender imbalances
produce more sexists. They cite Kant’s distinction: we may not fail in our
duties to AIs, but we may very well fail with regard to them. Thus,
Klonschinski and Kühler’s piece draws a nice contrast with Stringer’s:
where he argues that our relationships with our pets can be enriching, our
relationships with artificial non-persons are potential minefields.

1.13   Thin Line Between Love and Hate

If there is any entity for which love is professed as much as love for another
human, it is one’s country. However, it is a love that is not always looked
on as a good thing. Erich Fromm’s sentiment is not uncommon:

Nationalism is our form of incest, is our idolatry, is our insanity. ‘Patriotism’

is its cult…Just as love for one individual which excludes the love for others
is not love, love for one’s country which is not part of one’s love for human-
ity is not love, but idolatrous worship. (Fromm 1955: 58)

However, in Chap. 14, “Patriotism and Nationalism as Two Distinct Ways

of Loving One’s Country,” authors Maria Ioannou, Martijn Boot, Ryan
Wittingslow, and Adriana Mattos (who themselves represent four different
nationalities) argue that not only can a principled distinction be made
between nationalism and patriotism, but that, while they are both instances
of love for one’s country, they are distinct kinds of love: only the former is
usually pernicious, and the latter does not necessarily prove a gateway to
it. Patriotism, which they define as love for one’s country along with a
sense of personal identification with it and concern for its well-­being, is
likened to the love a child has for their parent. Nationalism, they allow, is
a more-contested term, but settle on the definition of nationalism used to
refer to European nationalism from the late nineteenth to mid-­twentieth
centuries. The difference between the two is that, while both feature
 1 INTRODUCTION 23

in-group love, only nationalism features out-group derogation, an obser-

vation that they back up citing both sociological literature and the results
of psychological experimentation. The love involved in nationalism has
this feature, they argue, because it is more akin to passionate love, which
carries with it both the refusal to acknowledge the flaws in the beloved
(which leads to jingoistic distorted evaluation of one’s own country above
others) and the possibility of loss and attendant desperation. Nationalists
are in love with a particular version of their country, one associated with
their particular ethnic or cultural group, and one that is easily threatened,
in a way that provokes the worst excesses that we see in nationalism. The
authors’ diverse disciplinary background makes this article stand out in
how it draws on a particularly wide ranging variety of literatures, psycho-
logical and sociological, along with literature and philosophy, to make a
compelling case for their distinctions. This is only fitting given how love
itself is the subject of so many disciplines, so it is refreshing to see a case be
made that empirical studies (like a study involving inhaling oxytocin prior
to running trolley-problem cases with in-group versus out-­group potential
victims) can elucidate our theoretical conclusions. If they are correct, their
analysis helps to explain what kind of national crises are most likely to
provoke nationalist violence, but also to rescue patriotism from the disre-
pute to which writers like Fromm have consigned it.

1.14   References
Badhwar, Neera Kapur. 1989. Friends as Ends in Themselves. In Eros, Agape and
Philia: Readings in the Philosophy of Love, ed. Alan Soble, 165–188. St. Paul,
MN: Paragon House.
Burns, Gregory. 2013. How Dogs Love Us: A Neuroscientist and His Adopted Dog
Decode the Canine Brain. New York: Houghton Mifflin Harcourt.
Dancy, Jonathan. 2004. Enticing Reasons. In Reason and Value: Themes from
Joseph Raz, ed. R. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael
Smith, 91–118. Oxford: Oxford University Press.
Dreifus, Claudia. 2019. Carl Safina Is Certain Your Dog Loves You. The New
York Times.
Fisher, Mark. 1990. Personal Love. London: Duckworth.
Franklin-Hall, Andrew, and Agnieszka Jaworska. 2017. Holding on to Reasons of
the Heart: Cognitive Degeneration and the Capacity to Love. In Love, Reason
and Morality, ed. Esther Kroeker and Katrien Schaubroeck, 20–38. New York:
Routledge.
Fromm, Erich. 1955. The Sane Society. New York: Rinehart and Winston.
24 S. CUSHING

Haslanger, Sally. 2012. Resisting Reality: Social Construction and Social Critique.
Oxford: Oxford University Press.
Hopwood, Mark, 2014. Love’s Work: Eros and Moral Agency. ProQuest
Dissertations and Theses.
———. 2017. “The Extremely Difficult Realisation That Something Other Than
Oneself Is Real”: Iris Murdoch on Love and Moral Agency. European Journal
of Philosophy 26: 477–501.
Hurka, Thomas. 2017. Love and Reasons: The Many Relationships. In Love,
Reason and Morality, ed. Esther Kroeker and Katrien Schaubroeck, 163–180.
New York: Routledge.
Kolodny, N. 2003. Love as Valuing a Relationship. Philosophical Review 112
(2): 135–189.
Krebs, Angelika. 2014. Between I and Thou – On the Dialogical Nature of Love.
In Love and Its Objects. What Can We Care For? ed. Christian Maurer, Tony
Milligan, and Kamila Pacovská, 7–24. Basingstoke: Palgrave Macmillan.
Mele, Alfred. 2019. Manipulated Agents: A Window to Moral Responsibility.
New York: Oxford University Press.
Nozick, Robert. 1990. Love’s Bond. In The Examined Life. Philosophical
Meditations, 68–86. New York: Simon & Schuster.
Rich, Adrienne. 1995. On Lies, Secrets, and Silence: Selected Prose 1966–1978. WW
Norton & Company.
Setiya, K. (2014). Love and the Value of a Life. Philosophical Review, 123(3),
p. 251–280.
Shpall, Sam. 2018. A Tripartite Theory of Love. Journal of Ethics and Social
Philosophy 13: 91–124.
Solomon, Robert C. 1994. About Love. Reinventing Romance for Our Times.
Indianapolis: Hackett Pub. Co., Reprint 2006.
Stocker, M. 1976. The Schizophrenia of Modern Ethical Theories. The Journal of
Philosophy 73 (14): 453–446.
Velleman, J. David. 1999. Love as a Moral Emotion. Ethics 109 (2): 338–374.
———. 2008. Beyond Price. Ethics 118: 191–212.
Williams, Bernard. 1981. Persons, Character and Morality. In Moral Luck, 1–19.
Cambridge: Cambridge University Press.
Wolf, S. 2004. The Moral of Moral Luck. In Setting the Moral Compass. Essays by
Woman Philosophers, ed. C. Calhoun, 113–127. Oxford, Oxford University Press.
Wynne, Clive. 2019. Dog Is Love: Why and How Your Dog Loves You. New York:
Houghton Mifflin Harcourt.
Zangwill, Nick. 2013. Love: Gloriously Amoral and Arational. Philosophical
Explorations 16 (3): 298–314.
 CHAPTER 2

Making Room for Love in Kantian Ethics

Ernesto V. Garcia

2.1   Introduction
Kant’s ethics is traditionally seen as defending an austere view of morality.
With his focus on moral duty and exceptionless universal laws, Kant seems
to leave out many important aspects of our moral lives, including personal
feelings like sympathy and compassion and more partial relationships like
friendship and love. As Bernard Williams writes in his classic work Ethics
and the Limits of Philosophy about this general outlook:

The important thing about morality is its spirit, its underlying aims, and the
general picture of ethical life it implies. In order to see them, we shall need
to look carefully at a particular concept, moral obligation […] Morality is
distinguished by the special notion of obligation it uses, and by the signifi-
cance it gives to it […] The philosopher who has given the purest, deepest,
and most thorough representation of morality is Kant. But morality is not
an invention of philosophers. It is the outlook, or, incoherently, part of the
outlook, of almost all of us. (Williams 2006: 174)

E. V. Garcia (*)
University of Massachusetts Amherst, Amherst, MA, USA
e-mail: evg@philos.umass.edu

© The Author(s), under exclusive license to Springer Nature 25

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_2
26 E. V. GARCIA

For Williams, Kant’s excessive preoccupation with moral duty above all
else reduces the moral agent to ‘a rational agent and no more’—that is, a
kind of ‘noumenal self, outside time and causality’ that we privilege at the
expense of ‘the concrete, empirically determined person that one usually
takes oneself to be’ (Williams 2006: 63).
Thus, it might be surprising for many readers to discover that Kant—as
well as many contemporary Kantian ethical approaches—in fact have a lot
to say about love in general.1 In this chapter, I try to answer three main
questions:

• What are Kant’s views about love? More specifically, what place, if
any, is there for love in Kant’s own ethical theory?
• What is the best way of thinking about love from a contemporary
Kantian perspective?
• How well does this broadly Kantian approach to love reflect our
ordinary intuitions on these matters?

This chapter has three parts. First, in “Kant’s Views About Love and
Friendship”, I examine Kant’s views about love. In particular, I discuss his
account of moral versus non-moral love as found throughout his various
writings and show how this closely parallels his account of moral versus
non-moral friendship. Second, in “Some Contemporary Kantian
Approaches to Love and Friendship”, I look at some recent Kantian
accounts of both friendship (Neera Kapur Badhwar) and love (J. David
Velleman), highlighting how they go beyond, and in many ways arguably
improve upon, Kant’s own views via their appeal to Kant’s Formula of
Humanity. Lastly, in “Conclusion: Assessing Kantian Moral Love”, I dis-
cuss the overall merits of what I call ‘Kantian moral love’ as found in all of
these different approaches. I argue that while Kantian moral love may cor-
rectly identify, from the moral point of view, how we ought to act and
think when loving other people, it fails to provide a complete account of
love, crucially leaving out certain key elements from the wide range of lov-
ing relationships we find ourselves in, especially romantic love. That is,
while Kantian moral love might offer us a morally ideal way to love other
people, it falls short of capturing the full essence of love—mainly because
love is not simply a moral affair but also a matter of the heart.

1
By ‘Kantian’, I mean approaches that are broadly inspired by, even though sometimes
substantively modifying or even rejecting, various aspects of Kant’s own ethical views.
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 27

2.2   Kant’s Views About Love and Friendship

What are Kant’s views about love? This issue is complicated by the fact
that Kant identifies two very different kinds of love, what I will call ‘moral’
versus ‘non-moral’ or ‘natural’ love.2 We find this distinction in nearly all
of major ethical writings. For Kant, non-moral or natural love—or what he
refers to as amor complacentiae (‘love as a passion’)—has three main fea-
tures. First, it is grounded in our sensible as opposed to rational nature.
Kant describes such natural love as an ‘inclination’, a ‘propensity of sensa-
tion’, and a ‘matter of feeling’.3 Second, it disposes us to help others. As
Kant puts it in Lectures on Ethics Collins, natural love amounts to ‘well-­
wishing from inclination’ or ‘well-doing […] arising from the heart’.4 And
third, because it is a merely contingent inclination or feeling and thus for
Kant not under our direct voluntary control, we cannot be morally
required to possess and/or display such natural love in general—in keep-
ing with Kant’s famous dictum that ‘ought implies can’.5
By contrast, moral love—or what Kant calls amor benevolentiae (‘love as
benevolence’)—also has three features that are by and large opposites of
non-moral love. First, it is grounded in our rational as opposed to sensible
nature—or, as Kant puts it in Lectures on Ethics Collins, ‘arises from prin-
ciples of the understanding’.6 Second and third, while moral love also
leads us to help others, it does not merely dispose but rather morally obli-
gates or requires us to do so. As Kant claims, moral love consists in ‘benev-
olence on principle’, ‘beneficence from duty itself’, ‘love from duty’,
‘well-doing from obligation’, and, more generally, a moral ‘command-
ment’ to be beneficent which ‘does not leave it to one’s discretionary
choice to make this one’s principle’.7 In the Metaphysics of Morals, Kant

2
Kant refers to such love as ‘pathological’ (pathologische), by which he means related to
pa-thos, that is, to our passive, sensible natures, as opposed to being abnormal or diseased.
In order to express Kant’s idea in a more neutral way, I have adopted the terminology ‘natu-
ral love’. For some helpful general discussions of Kant on love and friendship, see Paton
1993, Baron 2002, and Sensen 2013.
3
Gr. 4: 399, KpV 5: 83, and MdS 6: 402. Note: All Kant references are to volume and
page numbers found in Immanuel Kants Schriften (Ausgabe der königlich preussischen
Akademie Wis-senshaften (Berlin: W. de Gruyter, 1902–)).
4
VE 27: 413.
5
See Gr. 4: 399, KpV 5: 83, MdS 6: 402 – cf. Rel. 6: 51 for Kant’s discussion of his famous
doctrine that ‘ought implies can’.
6
VE 27: 413.
7
VE 27: 413. For a Kantian account of love which rejects Kant’s own emphasis on benefi-
cence, see Ebels-Duggan 2008.
28 E. V. GARCIA

identifies what he calls our ‘duties of love’ as beneficence, gratitude, and

sympathy, where by the latter he does not mean a merely passive or recep-
tive feeling that we cannot directly control, but rather an active willingness
to be open to and share in the feelings of others.8
Kant offers a concise summary of this overall distinction between what
I am calling moral versus non-moral or natural love in the Groundwork,
writing:

It is in this way, no doubt, that we are to understand the passages from

Scripture that contain the command to love one’s neighbor, even our enemy.
For love as inclination cannot be commanded, but beneficence from duty
itself—even if no inclination whatsoever impels us to it, indeed if natural and
unconquerable aversion resists—is practical and not pathological love, which
lies in the will and not in the propensity of sensation, in principles of action and
not in melting compassion; and only the former can be commanded. (empha-
sis added)9

Notably, we find a parallel distinction in Kant’s treatment of friendship.

Following Aristotle, Kant distinguishes between three basic kinds of
friendship: two non-moral kinds of friendship, namely, (1) friendship
based on need and (2) friendship based on taste or what Kant calls ‘aes-
thetic friendship’ and (3) friendship based on a moral attitude, or moral
friendship. While this distinction involves many similar contrasts—for
example, non-moral friendship is based on feelings, whereas moral friend-
ship is based on reason; non-moral friendship cannot be morally required
of us, whereas moral friendship can, and so forth—moral friendship differs
from moral love for Kant in one important respect, namely, it incorporates
elements of both love and respect. In the Metaphysics of Morals, Kant draws
a famous analogy between the two fundamental forces of repulsion and
attraction in our physical world and the two fundamental principles that
govern our ‘moral world’, namely, respect which makes us ‘keep our dis-
tance’ from other people and love which ‘brings us closer to them’.10 In
general, moral friendship is more complex than moral love insofar as it
involves both forces. That is, as Kant writes, moral friendship is ‘the union
of two persons through equal mutual love and respect’.11 For Kant, moral

8
MdS 6: 452–479.
9
Gr. 4: 399.
10
MdS 6: 469.
11
MdS 6: 470.
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 29

friendship requires not only being lovingly concerned about and sharing
in each other’s well-being but also respecting the autonomy of the other
person, all of this performed on mutually equal terms. Nevertheless,
despite this key difference, the main parallel remains the same. In general,
Kant sees a stark divide between moral and non-moral versions of love and
friendship, a highly dualistic approach which reflects the more fundamen-
tal division between reason and our empirical nature in his broader ethi-
cal theory.

2.3   Some Contemporary Kantian Approaches

to Love and Friendship

In this section, I discuss some recent Kantian approaches to friendship

(Badhwar) and love (Velleman). What is most striking about these
approaches is that they largely ignore Kant’s official views about love and
friendship, including his distinction between moral and non-moral ver-
sions of both phenomena. Instead, they notably take as their starting point
an attractive feature of Kant’s ethics that we have not yet discussed: namely,
Kant’s well-known ‘Formula of Humanity’ (FH). According to FH, we
ought to ‘treat humanity, in your own person as well as the person of any
other, always at the same time as an end, never merely as a means’.12
What does Kant mean by FH? To explain this, we need to answer two
questions: (i) what does it mean to treat ourselves or other people as what
Kant calls an ‘end-in-itself’ as opposed to a ‘mere means’?; and (ii) what
exactly does Kant mean by ‘humanity’? First, what does it mean to treat
somebody as an ‘end-in-itself’ versus as a ‘mere means’? To take a con-
crete example, consider how we treat something like, say, a tool such as a
hammer, as a ‘mere means’. There are at least three basic attitudes involved.
In general, we regard this object as:

1a. having merely instrumental value, that is, we value it solely for what it
can achieve or do for us
1b. replaceable/fungible, that is, as simply interchangeable with any other
equivalent thing
1c. merely conditionally valuable, that is, its value varies depending on
changing external circumstances (e.g., we would regard it as no longer
valuable if, say, we have already achieved or given up our original end)
12
Gr. 4: 429.
30 E. V. GARCIA

By contrast, Kant insists that we ought to treat ‘ends-in-themselves’ in

an exactly opposite way. That is, we should regard them as:

2a. having non-instrumental or final value, that is, we value them simply
for their own sake as opposed to what they can achieve or do for us
2b. irreplaceable/non-fungible, that is, as somehow special or distinctive in
its own right as opposed to something we would simply treat as
exchangeable for some equivalent thing
2c. unconditionally valuable, that is, where it has an intrinsic value which
stays the same regardless of changing external circumstances

Second, what does Kant mean by ‘humanity’? In general, Kant identi-

fies ‘humanity’ with our rational nature or, more specifically, our rational
capacity to set and pursue ends for ourselves.13 On one traditional con-
strual—as represented above in Williams’ reading of Kant—‘humanity’
refers to our bare abstract rational nature to the exclusion of all those
particular traits which distinguish us as unique individuals. Seen this way,
critics like Robert Noggle and Robin Dillon argue that Kant actually
‘ignores’ persons in his moral theory insofar as a person is ‘much more
than a mere instance of rational agency’ but a ‘being with a particular life,
a particular psychology, and a particular set of attachments, goals, com-
mitments, and so on’ (Noggle 1999: 454) and so should be treated not
merely ‘as instances of generic personhood, but as the whole fully specific
person she is’, taking into account ‘precisely [those] contingencies that
make me who I am’ (Dillon 1992: 74).14 By contrast, both of the thinkers
discussed later reject this traditional construal, instead insisting that treat-
ing people as ‘ends-in-themselves’ not only allows for, but indeed requires,
seeing other people in their full particularity.
To summarize: In unpacking Kant’s FH, we have seen that treating the
‘humanity’ of other people as ‘ends-in-themselves’ involves regarding
their rational agency as being non-instrumentally and unconditionally
valuable as well as in some sense irreplaceable, and acting accordingly. In
the rest of this section, I discuss how Badhwar and Velleman directly apply
Kant’s FH to analyze friendship and love. More specifically, I want to
explore the following question: How does viewing friends and people we
love as Kantian ‘ends-in-themselves’ transform the way we think about the

13
See Gr. 4: 428, Rel. 6: 26, and MdS 6: 392.
14
For a detailed defense of Kant on these matters, see Garcia 2012.
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 31

nature of friendship and love in general? In “Conclusion: Assessing Kantian

Moral Love”, I conclude by critically evaluating the respective merits of
Kant’s own view vis-à-vis these more contemporary Kantian approaches to
friendship and love.
In her classic article ‘Friends as Ends in Themselves’, Neera Kapur
Badhwar straightforwardly claims that we should regard friends as Kantian
‘ends-in-themselves’. What does this amount to? Badhwar argues that in
an ideal moral friendship, we should not view our friends as ‘mere means’
related to those ‘incidental features that make her useful or pleasurable’
(Badhwar 1989: 166). Instead, we should see them as having ‘an intrinsic,
not instrumental value’ as well as being ‘a unique and irreplaceable indi-
vidual’ on ‘account of what she essentially is’ (Badhwar 1989: 165),
directly corresponding to (2a) and (2b). Further, our love for our friends
should not vary relative to merely external changing circumstances. As
Badhwar puts it, following Shakespeare, our love for our friends ‘is not
love if it alters whenever it alteration finds’ (Badhwar 1989: 169), thus cor-
responding to (2c). However, unlike what she takes to be the type of
‘unconditional’ agapeistic love which loves someone totally independently
of their specific qualities—say, merely in virtue of their being a ‘Human
Being or Instance of (some F)’, for example, a rational agent—Badhwar
maintains that the true object of our love should be the ‘unique’, ‘irre-
placeable’, and ‘historical’ individual (Badhwar 1989: 169).
In his highly influential discussion ‘Love as a Moral Emotion’, J. David
Velleman likewise argues that truly loving people involves seeing them as
Kantian end-in-themselves. Velleman rejects traditional accounts of love as
a mere feeling, à la non-moral Kantian love, or as a drive or desire, à la
Freud and many contemporary analytic philosophers. Instead, Velleman
sees love as a special mode of attention and valuation (Velleman 2006: 76,
85–86). In particular, love consists in regarding our beloved—in an explic-
itly Kantian sense—as an ‘end’, where this amounts to ‘awareness of a
value inhering in’ the person herself (Velleman 2006: 94). For Velleman,
this involves valuing other people in a non-instrumental manner, not as
‘an independent aim to be realized’ but rather as something that we sim-
ply ‘act for the sake of’ (Velleman 2006: 89–90), in keeping with (2a). In
addition, Velleman thinks that we should regard the beloved as having
‘dignity’ or ‘incomparable value’ in Kant’s sense, where what makes some-
one ‘truly irreplaceable is a value that commands appreciation for it as it is
in itself, without comparison to anything else, and hence without substitu-
tions’ (Velleman 2006: 99–100), in keeping with (2b).
32 E. V. GARCIA

Lastly, to appreciate how Velleman incorporates (2c) in his account of

love, we need to look more closely at how he thinks about the object of
love, that is, the beloved. With Kant’s FH obviously in mind, Velleman
claims that the true object of love is the other person’s ‘humanity’ or their
‘rational nature’ (Velleman 2006: 100). However, by ‘rational nature’,
Velleman does not mean just our intellects, but rather our ‘capacity of
appreciation or valuation’ more broadly, or put simply, ‘the core of our
reflective concern’ in terms of ‘our capacity to love’ itself (Velleman 2006:
100). In this way, loving another person amounts to our heart ‘respond[ing]
to […] another heart’ (Velleman 2006: 100). Nonetheless, while our love
is grounded in the ‘universal value’ that the beloved has simply in virtue of
‘being a person’ or an ‘instance of rational nature’, the way we necessarily
come to respond to this value is by loving their concrete ‘empirical per-
sona’. This persona can include various ‘observable features’ about them
like the way they talk or walk, or even how they wear their hat or sip their
tea, which we take to be ‘external symbols or reminders’ of their internal
value as persons which they retain regardless of any changing external
circumstances (Velleman 2006: 100, 106), thus in keeping with (2c).

2.4   Conclusion: Assessing Kantian Moral Love

What should we think about what I will broadly call ‘Kantian moral love’,
where this is meant to include both Kant’s account of moral love spelled
out in terms of ‘duties of love’ and contemporary Kantian approaches
which see love as a ‘moral emotion’? In particular, how does (1) Kant’s
official historical view, which insists upon a sharp dichotomy between
feeling-­based ‘non-moral’ versus reason-based ‘moral’ friendship and love,
measure up against (2) various contemporary Kantian approaches, accord-
ing to which both friends and people we love are best understood as
Kantian ‘ends-in-themselves’ in terms of being non-instrumentally and
unconditionally valuable and in some sense irreplaceable or special in their
own right?
On the one hand, I think that contemporary Kantian approaches sig-
nificantly improve upon Kant’s original views in two main ways. First,
unlike Kant’s approach, they appeal to Kant’s FH to analyze the nature of
love. This is a major advantage not only because many commentators
regard FH as one of the most appealing aspects of Kant’s ethical theory,
insofar as it identifies a highly attractive moral ideal—namely, treating peo-
ple as ends-in-themselves with a fundamental dignity—that many seem to
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 33

think underlies our modern-day conceptions of basic human rights. In

addition, it also puts the emphasis in the right place when it comes to
Kantian moral love. It turns our attention away from what Kant himself
focuses on in his own account of moral love, namely, our acting from
moral principle in terms of fulfilling our ‘duties of love’ such as benefi-
cence and gratitude. Instead, it highlights what seems to be the true object
of moral love, namely, people themselves understood in terms of their basic
‘humanity’.
Second and relatedly, such contemporary Kantian approaches are much
more akin to how we ordinarily think about love. In contrast to the type
of universalistic love involved with Kant’s moral duties of love that we owe
to all people in general—which might be more suitably called ‘philan-
thropy’, that is, ‘love of humanity’ in a very literal sense—these contem-
porary Kantian approaches instead demand that we attend to the
individuality of those we love. By focusing on the beloved as a ‘unique’,
‘irreplaceable’, and ‘historical’ individual (Badhwar 1989: 69) or in terms
of their concrete ‘empirical persona’ and the various ways that we ‘respond
to [their] value through that persona’ (Velleman 2006: 107), such con-
temporary approaches to love—unlike Kant’s own view—instruct us to
embrace the full concrete historical particularity of those we love, includ-
ing our special relationships with them.
On the other hand, I think that there is something deeply problematic
about such contemporary Kantian approaches if they claim to offer a full
analysis of the essence of love as such. We can see this most clearly in
Velleman’s synopsis of his view, when he writes: ‘All that is essential to love,
in my view, is that it disarms our emotional defenses towards an object in
response to its incomparable value as a self-existent end’ (Velleman 2006:
99, emphasis added). What should we think about Velleman’s claim that
what is ‘essential’ to love is how it emotionally disarms us in response to
some incomparably valuable end-in-itself? The main problem is that
Velleman’s account—which is open to two very different interpretations—
seems to face a fundamental dilemma.
On one horn of the dilemma, related to what I will call a ‘weak reading’
of Velleman, we can interpret him as merely claiming that, in love, (1) we
necessarily make ourselves emotionally vulnerable in response to another
person and (2) if Kant’s ethical views are true, then we are directly
responding to a person who is end-in-itself with incomparable value
regardless of whether we are actually aware of this or not. Both (1) and (2)
seem correct, perhaps even trivially so. The basic worry is that, taken
34 E. V. GARCIA

together, they do not establish Velleman’s substantive thesis that (3) love
is a ‘moral emotion’. In fact, it (1) seems to be just a non-moral truism
about love (viz., that in love, we invariably make ourselves emotionally
vulnerable in response to another person), whereas (2) is just a doctrinal
truism about Kant’s ethics (viz., that if Kant’s view is true, then people we
love are ends-in-themselves with incomparable value). However, affirming
the mere conjunction of the non-moral truism related to (1) and the moral
truism related to (2) in no way establishes Velleman’s quite substantive
thesis that (3) love is an inherently moral emotion.
On the other horn of the dilemma, related to what I will call a ‘strong
reading’ of Velleman, he is instead claiming that, in love, (1’) we make
ourselves emotionally vulnerable in response to another person and (2’)
this response necessarily involves our awareness of the other person as an
end-in-itself with incomparable value. This reading squares better with
Velleman’s own description of both respect and love, where he argues that
the former involves ‘awareness of a value that arrests our self-love’, while
the latter involves ‘awareness of a value that arrests […] our tendencies
toward emotional self-protection from another person’—where the ‘value’
that we are aware of in both cases, as discussed earlier, is the ‘universal
value’ of our ‘humanity’ understood as ‘a capacity of appreciation or valu-
ation’ (Velleman 2006: 95, 100–101). While I think that this interpreta-
tion fits much better with Velleman’s text than the weak reading, the main
problem is that it seems to involve a highly implausible account of what is
‘essential’ to love. Put differently, Velleman fails to identify either neces-
sary or sufficient conditions for love here.
First, Velleman fails to identify sufficient conditions for love. That is,
even if we fulfill conditions (1’) and (2’), this still does not guarantee that
we are actually loving the person in question. To give just a few examples,
it seems possible that, à la Velleman, (1’) I make myself emotionally vul-
nerable to X and (2’) that I recognize and respond to X’s value as an ‘end-­
in-­itself’ based on their ‘capacity of appreciation or valuation’, but I do
not actually love X but am instead merely:

a) making an appeal to X’s fellow humanity to provide aid when I

am in need
b) asking for X to show mercy and refrain from harming me because X is
an aggressor
c) soliciting X to join me in some worthwhile moral cause
d) simply liking X as a person and enjoying spending time with them
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 35

In all these cases, I meet Velleman’s conditions insofar as (1’) I make

myself emotionally vulnerable to X with all the attendant potential joys
and sorrows and (2’) I am aware of, and, indeed, my response to X presup-
poses, that X shares a common invaluable humanity with me on the basis
of which it makes sense for me to even morally appeal to them in the first
place. Nevertheless, it seems like none of these scenarios—asking for X’s
help, petitioning X to show mercy to me, hoping for X’s solidarity, or
merely liking X as a person (where it seems too extreme to subsume all
cases of liking another person to loving them)—necessarily requires that I
actually love the other party in question.
Second, Velleman fails to identify necessary conditions for love. That is,
it seems possible for us to love somebody without satisfying both condi-
tions (1’) and (2’). Take, for example, A romantically loving B. I think
Velleman is right that this typically involves (1’): in virtue of romantically
loving B, A makes themselves emotionally vulnerable to B. However, (2’)
does not seem necessary here. That is, it seems fully possible for A to
romantically love B while not appreciating or ever even being aware of the
fact that B has ‘incomparable value’ as a kind of Kantian ‘end-in-itself’.15
In the end, to partially defend Kant’s original views about love and
friendship, I think it is important to retain at least some version of Kant’s
distinction between moral versus non-moral friendship and/or love. That
is, I think we should avoid overly moralized accounts of love and friend-
ship that exclude from the outset the very possibility of non-moral ver-
sions of each phenomenon. Indeed, Kant himself insists that what we are
calling non-moral or natural love still has an important place in our human
lives, even describing it at times as ‘very beautiful’, ‘amiable’, and deserv-
ing of our ‘praise and encouragement’ although falling short of full-­
fledged ‘moral worth’.16 To her credit, Badhwar recognizes this in her
Kantian analysis of friendship. While she praises what she calls moral ‘end
friendship’ as ‘the best, most complete’ kind of friendship, she nonetheless
affirms, following Aristotle, that there exist many different kinds of valu-
able non-moral friendships, too (Badhwar 1989: 165–166).
By contrast, Velleman’s account falls prey to this pitfall. His overly mor-
alized Kantian account of love as a ‘moral emotion’—where love

15
Cf. Berit Brogaard’s claim that it seems we can romantically love somebody ‘for any
reason or no reason at all’ and that such love does not ‘require us to appreciate the positive
value of an-other person’ (Brogaard 2015: 73).
16
See Gr. 4: 398 and KpV 5: 82.
36 E. V. GARCIA

necessarily involves being aware of and responding to the value of another

person’s humanity in the Kantian sense—fails to capture what is essential
to all cases of love. Seen in this light, I think that Berit Brogaard’s remarks
about romantic love in contrast to what she calls ‘companionate love’—
where the latter is closely akin to Velleman’s view—are instructive here. As
Brogaard writes:

I don’t think there’s a single kind of true love, and I fully believe that roman-
tic love is real love. It’s as real and true as the love you feel for your grand-
father or your childhood friend. Granted, it’s different from companionate
love and attachment love, but it is, nonetheless, love (Brogaard 2015: xii,
first emphasis added)

At the end of the day, I think we should recognize that there can be
many different kinds of love, both moral and non-moral in nature. To bor-
row Kant’s terminology from his account of the ‘highest good’ in the
second Critique,17 Kantian moral love represents a ‘supreme good’ (supre-
mum)—that is, the best or ideal form of love between people, at least from
a moral perspective—contra Velleman, it does not, and indeed as Kant
himself insists, cannot, represent the ‘complete good’ (consummatum)—
that is, the full range or essence of our loving human relationships in
general.18

2.5   References
Badhwar, Neera Kapur. 1989. Friends as Ends in Themselves. In Eros, Agape, and
Philia: Readings in the Philosophy of Love, ed. Alan Soble, 165–188. St. Paul,
MN: Paragon House.
Baron, Marcia. 2002. Love and Respect in the Doctrine of Virtue. In Kant’s
Metaphysics of Morals: Interpretive Essays, ed. Mark Timmons, 391–407.
Oxford: Oxford University Press.
Brogaard, Berit. 2015. On Romantic Love: Simple Truths About a Complex
Emotion. Oxford: Oxford Press.
Dillon, Robin. 1992. Care and Respect. In Explorations in Feminist Ethics: Theory
and Practice, ed. Eve Browning Cole and Susan Coultrap-McQuin, 69–81.
Bloomington: Midland Press.

17
KpV 5: 110.
18
I would like to thank both Simon Cushing and Nicholas Vallone for their extremely
helpful and insightful feedback on this chapter.
 2 MAKING ROOM FOR LOVE IN KANTIAN ETHICS 37

Ebels-Duggan, Kyla. 2008. Against Beneficence: A Normative Account of Love.

Ethics 119: 142–170.
Garcia, Ernesto V. 2012. A New Look at Kantian Respect for Persons. Kant
Yearbook 4: 69–90.
Kant, Immanuel. 1996. Metaphysics of Morals (MdS). Cambridge: Cambridge
University Press.
———. 1997a. Critique of Practical Reason (KpV). Cambridge: Cambridge
University Press.
———. 1997b. Groundwork of the Metaphysics of Morals (Gr). Cambridge:
Cambridge University Press.
———. 1997c. Lectures on Ethics (VE). Cambridge: Cambridge University Press.
———. 1998. Religion Within the Boundaries of Mere Reason (Rel). Cambridge:
Cambridge University Press.
———. 1997. Love and Solipsism. In Love Analyzed, ed. Roger Lamb, 123–152.
Boulder, CO: Westview Press.
Noggle, Robert. 1999. Kantian Respect and Particular Persons. Canadian Journal
of Philosophy 29: 449–478.
Paton, H.J. 1993. Kant on Friendship. In Friendship: A Philosophical Reader, ed.
Neera Kapur Badhwar, 133–154. Ithaca: Cornell University Press.
Sensen, Oliver. 2013. Friendship in Kant’s Moral Thought. In Thinking About
Friendship: Historical and Contemporary Philosophical Perspectives, ed. Damian
Caluori, 143–160. New York: Palgrave Macmillan.
Velleman, J. David. 2006. Love as a Moral Emotion. In Self to Self: Selected Essays,
70–109. New York: Cambridge University Press.
Williams, Bernard. 2006. Ethics and the Limits of Philosophy. New York:
Routledge Press.
 CHAPTER 3

Iris Murdoch and the Epistemic Significance

of Love

Cathy Mason

3.1   Introduction
In The Sovereignty of Good (1970), Iris Murdoch gives love an intellectual
and epistemic standing with which many philosophers would be uncom-
fortable. She says not only that it is epistemically valuable—a claim already
too strong for many, given the lover’s seeming tendency to misperceive1—
but also that we do not see reality as it truly is unless we love. This is a
puzzling claim. We tend to think that the very point of objective knowl-
edge is to abstract away from any personal, particular point of view, taking
something like what Bernard Williams (1978) calls ‘the absolute
conception’ as our standard. And we often think of love as a paradigm of
just such a personal, particular—and perhaps distorted—point of view. It
thus seems precluded from playing the epistemic role that Murdoch
assigns to it. Part of my aim in the present chapter is to offer an

1
Keller (2004) and Stroud (2006), for example, suggest that friendship constitutively
involves epistemic partiality.

C. Mason (*)
Wadham College, University of Oxford, Oxford, UK

© The Author(s), under exclusive license to Springer Nature 39

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_3
40 C. MASON

interpretation of the conception of love in The Sovereignty of Good such

that this seeming tension between love and objective knowledge is
dissolved.2
For Murdoch, love has a particular significance in the perception of
moral reality. There is an obvious causal connection between love and
morality: love can be a powerful factor in motivating us to act in morally
admirable ways. However, Murdoch’s claim is that love is also epistemi-
cally significant for our ethical lives. On the Murdochian interpretation,
‘loving thy neighbor’ entails not only being motivationally affected by
one’s neighbor’s well-being, but also standing in an epistemic relation to
them that involves knowledge and continuous progression toward truer
understanding of them. This may seem counterintuitive, but I will suggest
that there are good reasons to take Murdoch’s account seriously. Her
claim is not ultimately as puzzling as it first appears.
I begin in “Murdoch’s Moral Framework” by outlining Murdoch’s
moral framework and the role of love within it. In “Alternative
Interpretations of Murdochian Love”, I then explore two contrasting
interpretations of Murdochian love proposed by Velleman (1999) and
Hopwood (2014, 2017) and discuss the ways in which each fails to do
justice to the full epistemic role Murdoch assigns to love. In “Love and
Realism”, I explore the notion of objectivity that underlies Murdoch’s
account. In “Love as a Virtue and a Perceptual Sensitivity”, I argue that
Murdochian love is best interpreted as a virtue, with a particularly lofty
position in the hierarchy of the virtues. This allows Murdoch’s claims
about love’s epistemic value to be understood while retaining her claims
about objectivity. My aims are not, however, only exegetical. In “The
Everyday Concept of Love: A Defense of Murdoch”, I argue that this
reconstructive exercise yields an illuminating account of our ordinary con-
ception of love.

2
In this chapter I shall focus on Murdoch’s conception of love in The Sovereignty of Good
(1970) and other early works: “Vision and Choice in Morality” (1956) and “The Sublime
and the Good” (1959). Her overarching ethical vision in later work such as Metaphysics as a
Guide to Morals (1992) is somewhat altered, becoming more heavily Platonic and mystical.
This corresponds with a linguistic change in Murdoch: in later work she refers primarily to
eros rather than love. There is thus reason to think that her conception of love may have simi-
larly developed and altered over time, and I shall not examine the later conception. I will use
the term ‘Murdochian love’ to refer only to the conception of love found in her early works.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 41

3.2   Murdoch’s Moral Framework

Murdoch’s discussion of love is framed by the conviction that there is an
essential kind of moral activity that is not reducible to publicly observable
actions. She calls this neglected kind of activity ‘attention’.3 According to
Murdoch, the dominance of broadly behaviorist theories of mind in ‘mod-
ern moral philosophy’ led to the conviction that “morality resides at the
point of action” (Murdoch 1970: 16) and that moral agency must there-
fore relate only to publicly observable outward action. Murdoch is wholly
resistant to this line of thinking, which she regards as distorting our under-
standing of what is at stake in ethics by disregarding important areas of our
ethical lives. Such a conception of morality automatically rules out phe-
nomena such as attention from moral consideration, but Murdoch claims
that these phenomena are deeply morally significant. She thus advocates
re-emphasizing the importance of various concepts that were peripheral in
contemporary moral philosophy, and I shall focus here on her attempted
reinstatement of the concept of love. Murdoch’s basic idea here is that we
must attend to objects, must see them in a morally significant way, before
we can hope for our publicly observable actions to be morally worthy.
More importantly, she insists that attention is itself a fundamentally moral
activity. Such attention, she thinks, is a kind of love.
Murdoch thus begins two of her most famous essays with assertions of
the significance of love in ethics. She claims that “love is a central concept
in morals” (Murdoch 1970: 2) and that “we need a moral philosophy in
which the concept of love, so rarely mentioned now by philosophers, can
once again be made central” (Murdoch 1970: 46). Having declared that
love is a central moral concept, she specifies that one role love fulfills is
epistemic: our coming to grasp moral truths, and the progressive deepen-
ing of our grasp of them, is a manifestation of love.
It is not obvious that love is morally and epistemically valuable. Love
can appear to be as much bound up with illusion as perception and to be
capable of leading to cruelty as well as self-sacrifice. (Think, for example,
of Othello’s claim after murdering Desdemona that he has “loved not
wisely but too well”.) Nonetheless, Murdoch presents love as fulfilling a
crucial ethical and epistemic role:

3
This is a term taken from Simone Weil (1956).
42 C. MASON

Love is the perception of individuals. Love is the extremely difficult realisa-

tion that something other than oneself is real. Love, and so art and morals,
is the discovery of reality. (Murdoch 1959: 51)

Love is knowledge of the individual. (Murdoch 1970: 28)

In particular situations ‘reality’ as that which is revealed by the patient eye of

love is an idea entirely comprehensible to the ordinary person.
(Murdoch 1970: 40)

Murdoch here describes the ordinary concept of love as having an epis-

temic dimension: it involves knowledge, discovery, or perception of the indi-
vidual and reality. Love is thus presented as fulfilling some kind of epistemic
role: it in some sense involves grasping truths.
Murdoch illustrates this epistemic role of love with an example that is,
for her, a paradigm case of both moral and epistemic progress through
loving attention:

A mother, whom I shall call M, feels hostility to her daughter-in-law, whom

I shall call D. M finds D quite a good-hearted girl, but while not exactly
common yet certainly unpolished and lacking in dignity and refinement. D
is inclined to be pert and familiar, insufficiently ceremonious, brusque,
sometimes positively rude, always tiresomely juvenile.
…Time passes, and it could be that M settles down with a hardened sense
of grievance and a fixed picture of D, imprisoned (if I may use a question-­
begging word) by the cliché: my poor son has married a silly vulgar girl.
However, the M of the example is an intelligent and well-intentioned per-
son, capable of self-criticism, capable of giving careful and just attention to
an object which confronts her. M tells herself: ‘I am old-fashioned and con-
ventional. I may be prejudiced and narrow-minded. I may be snobbish. I am
certainly jealous. Let me look again.’ Here I assume that M observes or at
least reflects deliberately about D, until gradually her vision of D alters….D
is discovered to be not vulgar but refreshingly simple, not undignified but
spontaneous, not noisy but gay, not tiresomely juvenile but delightfully
youthful, and so on. (Murdoch 1970: 17–18)

Murdoch describes M’s transition here as a transition to viewing D

‘lovingly’. Although not attended by any outward change in M’s behavior,
it is intended by Murdoch to be a fundamentally moral transition, one in
which M’s moral standing improves. As M lovingly attends to D, she
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 43

becomes able to perceive features of D that were previously obscured or

distorted by latent selfishness and prejudice. For example, as M attends
lovingly to D, D’s delightful youthfulness, which was previously obscured
by M’s snobbery and jealousy, becomes discernible to M.
Murdoch’s claim is thus that love fulfills an epistemic role: love involves
attending to reality, and results in a deepening understanding of reality.
She understands attending to reality not merely as something that one can
do lovingly but as itself “an exercise of love” (Murdoch 1970: 42). Such
loving attention, she suggests, will progressively lead one toward a deeper,
more adequate conception of reality: “[w]hen M is just and loving she sees
D as she really is” (Murdoch 1970: 37). In the next section I shall explore
two interpretations of Murdoch that offer ways of spelling out how
Murdochian love performs this function. I will suggest that both fall short
of Murdoch’s own claims about love’s epistemic significance.

3.3   Alternative Interpretations

of Murdochian Love

Some philosophical discussions of love have assumed that there is a ten-

sion between morality and love.4 By contrast, both Velleman (1999) and
Hopwood (2014, 2017) follow Murdoch and use her thought in articu-
lating models of love that aim to vindicate its moral significance and that
make some place for love’s epistemic significance. However, both square
Murdoch’s thought that love is a moral activity with her thought that love
is epistemically rich only at the expense of denying further aspects of her
view; they connect love and morality but fail to explain the broader con-
nection between love and knowledge. An alternative account of love is
thus needed to elucidate its role in Murdoch’s thought. I will consider
Velleman and Hopwood’s accounts in turn, before discussing Murdoch’s
background commitment to realism in “Alternative Interpretations of
Murdochian Love” and offering my own account of Murdochian love in
“Love and Realism”.

4
See, for example, Williams’ (1981) discussion of the permissibility of saving one’s wife
rather than a stranger, suggesting that the demands of close relationships might conflict with
those of morality.
44 C. MASON

In an account of love that he aligns with Murdoch’s, Velleman argues

that love is a rational state capable of being justified by reasons.5 On his
account, love constitutes an appreciation of inherent value in the beloved
which brings with it an emotional vulnerability to them. Specifically, he
regards love as involving an “arresting awareness” (Velleman 1999: 360)
and appreciation of the value of rational natures and therefore suggests
that Murdochian love resembles Kantian respect.6 He claims that respond-
ing to the value of rational natures with respect is a rationally ‘required
minimum’ and that responding with love is an ‘optional maximum’. This
is an epistemically rich account of love insofar as love is understood by
Velleman as constituting a recognition of value.
However, this cannot be Murdoch’s conception of love. Velleman sug-
gests that love is a morally and rationally optional response to the uncon-
ditional value of rational natures, but Murdoch contends that love is
morally necessary. In response to Kant’s contention that only practical
love (performing loving actions) can be a duty, Murdoch argues that
‘pathological love’ (love as an affective state or quality of consciousness)
also matters morally: “I do not agree that only practical love can be com-
manded….Pathological love can be commanded too, and indeed if love is
a purification of the imagination, must be commanded” (Murdoch 1959:
55). For Murdoch, what we are morally ‘commanded’ to do extends far
beyond publicly observable actions. Her claim is that we are also obliged to
love in the sense of lovingly attending to others (‘purifying the imagina-
tion’). For Murdoch, an unloving perspective will simply not allow one to
perceive truths about the world that the lover can see, and therefore lov-
ingly attending to the world is both epistemically and morally obligatory.
Moreover, on Velleman’s account, love does not reveal the features of
persons that Murdochian love reveals. For Murdoch, love is an acknowl-
edgment of the reality of particulars outside oneself, whereas, for Velleman,
it is directed at the same universal aspect of each person, their rational
nature.7 Murdoch explicitly criticizes Kant for exactly this failure:

5
For example, Velleman writes that “[t]his hypothesis would explain why love is an exer-
cise in ‘really looking’, as Murdoch claims” (Velleman 1999: 361)
6
Bagnoli (2003) also suggests that Kantian respect and Murdochian love are “significantly
analogous”: they “exhibit a similar phenomenology and work likewise, as constraints on
deliberation” (Bagnoli 2003: 506, 485).
7
Clarke (2012) emphasizes the idea that Murdochian attention involves seeing an object
“in all of its (significant) particularity”, and the political potential of this idea for overcoming
prejudice (Clarke 2012: 238).
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 45

Kant does not tell us to respect whole particular tangled-up individuals, but
to respect the universal reason in their breasts. In so far as we are rational
and moral we are all the same, and in some mysterious sense transcendent to
history. (Murdoch 1959: 51)

Given Velleman’s belief in the similarity between Murdochian love and

Kantian respect, it is unsurprising that this criticism of Kantian respect can
equally be applied to Velleman’s account of love. For Murdoch, love
directs one’s attention “towards the great surprising variety of the world”
(Murdoch 1970: 66). On the Murdochian account, loving another person
involves directing one’s attention toward a particular concrete individual,
not simply toward a universal abstract property instantiated in them.
Murdoch’s emphasis on particularity thus precludes Velleman’s account
from capturing Murdochian love.8
Hopwood’s account, on the other hand, correctly emphasizes the par-
ticularity of Murdochian love. He contends that Murdoch’s conception of
love is that of Platonic eros, which he understands as follows:

[E]ros is (i) a form of desire that is (ii) directed at a particular object whose
value (iii) cannot be captured under a closed description, that (iv) engages
the imagination, and that (v) carries with it the awareness of a normative
demand on the subject. (Hopwood 2014: 61)

A closed description, for Hopwood, is one in which the object can be

exhaustively characterized in terms of its properties, where one’s relation
is to any object that falls under the relevant description.9 To desire an
object whose value cannot be captured in a closed description is therefore
to value an irreducibly particular object. According to Hopwood, eros is a
form of desire, an affective state that nonetheless involves recognition of
the capacity of the beloved, as such, to place demands on oneself, enabling
one to see one’s response to the object of love as potentially inadequate or
as falling short in some way. Hopwood thus claims that love has an

8
Murdoch does believe that ‘the Good’ is also an object of love, which appears to be in
tension with this. However, she maintains that it is a ‘concrete universal’ (Murdoch
1970: 29).
9
Hopwood illustrates the idea of loving someone under a closed description with the fol-
lowing example: “If we were to propose to take Romeo away and replace him with another
person possessing exactly the same set of characteristics…Juliet would presumably not be
happy to accept the swap. Her desire for Romeo is a desire for a particular individual, and
precisely because of this, the value that she sees in him cannot be captured under a closed
description” (Hopwood 2014: 8).
46 C. MASON

epistemic component, the potential to be aware of one’s response as fail-

ing to do justice to its object.10
The idea that the object of love cannot be captured under a closed
description introduces into this account an ineliminable particularity. On
this account, love is not a response to an abstract or universal property, but
to a particular person or object. Moreover, on Hopwood’s account, love
plays both moral and epistemic roles, since it is understood as a form of
desire that brings with it an awareness of normative demands upon
the agent.
However, Hopwood’s account does not do full justice to the epistemic
role Murdoch assigns to love. Murdoch contends that love is important
not only in the perception of normative demands that loving awareness of
objects places on the lover, but in the perception of objects themselves.
Love, for Murdoch, primarily reveals objects themselves rather than nor-
mative demands that agents face. She claims that “love is knowledge of the
individual” (Murdoch 1970: 28): loving attention is necessary for any
truly adequate perception of a person, object, or situation itself, not merely
knowledge of normative demands. Hopwood’s account of Murdochian
love is therefore, like Velleman’s, too narrow to account for the fullness of
the epistemic role Murdoch assigns to it.

3.4   Love and Realism

One possible way of affording love the broad epistemic role Murdoch
insists upon would be to understand moral reality as constitutively depen-
dent upon the subject. This is suggested by Hopwood’s claim that it is in
virtue of loving the object that one sees it as making demands on oneself.
One way of reading this would be as suggesting that there is no fixed
moral reality for the observer to respond to that exists apart from the
observer’s perception and their love. On this reading, the reality that
places demands upon the observer does so, at least in part, because it is
loved, not vice versa. Thus interpreted, Hopwood suggests that Murdoch
should not be read as subscribing to a robust form of realism. If the claim
that love enables us to get to the objective truth about the beloved is relin-
quished, the apparent tension between love (a seemingly personal and

10
Hopwood depends heavily on Murdoch’s later work, particularly Metaphysics as a Guide
to Morals (1992). Although Hopwood’s account does not capture Murdoch’s early concep-
tion of love, it may capture her later conception of it.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 47

particular state) and knowledge (apparently objective) disappears. If the

objectivity of moral reality is given up, there can be no tension between it
and the epistemic role Murdoch assigns to love: love reveals moral reality
because that reality is (at least partly) constituted by its being loved.
However, Murdoch is committed to the idea that the truth of moral
claims is constitutively dependent on reality outside the observer’s percep-
tions and beliefs. M, for example, is described as ‘discovering’ D’s moral
qualities, suggesting that such features do not depend on her. Murdoch
also describes M’s loving re-evaluation of D as revealing her “as she really
is” (Murdoch 1970: 37), suggesting that the moral evaluation depends on
D’s characteristics, rather than on M.
Pervasive throughout The Sovereignty of Good is the image of the moral
life as an exercise of vision. Importantly, Murdoch conceives of this vision
as revealing what is there independently of the perceiver’s conception of it
or attitude toward it. In her discussion of the arts as introductions to (and
initial participations in) the moral life, Murdoch claims that what is
required of both is “unsentimental, detached, unselfish, objective atten-
tion” (Murdoch 1970: 64). Elsewhere, she focuses on the connection
between the real and the true: “the realism (ability to perceive reality)
required for goodness is a kind of intellectual ability to perceive what is
true” (Murdoch 1970: 64). In fact, in her criticism of the ‘current view’
of persons, she states that “we have lost the vision of a reality separate from
ourselves” (Murdoch 1970: 46). Love, she suggests, helps us to discern
what is true, a reality that is separate from ourselves. Murdoch therefore
cannot be suggesting that love reveals moral reality in virtue of moral real-
ity being dependent upon the perceiver’s loving stance.11
Indeed, Murdoch suggests that ‘fantasy’, the projections of one’s own
self in one’s view of the world, is the “chief enemy of excellence in moral-
ity” (Murdoch 1970: 59). For her, projections of the self in one’s vision of
the world are fundamentally distorting. It is by directing one’s attention
away from the self and the distorting fantasies generated by the selfish ego
that moral reality is progressively revealed. On her account, moral reality
thus cannot constitutively depend on the subject.
How, then, can we make sense of the idea that loving attention reveals
a reality that is separate from the observer? I want to suggest that
Murdoch’s claim is best understood in the context of her repudiation of

11
For more on Murdoch’s metaethics, see Jordan (2014). He understands her as a realist
committed to cognitivism, success theory, and objectivism.
48 C. MASON

the idea that objective reality is that which is revealed by value-neutral

perception. Rather, she claims that all perception itself is morally imbued.12
The moral realist, Murdoch argues, ought not to attempt to strive to dem-
onstrate the objectivity of morality through its assimilation into the ‘hard’
world of impersonal facts that purport to be ‘neutral’ and available to
anyone, but instead should reject such a model of objectivity:

[G]oodness is connected with knowledge: not with impersonal quasi-­

scientific knowledge of the ordinary world, whatever that may be, but with
a refined and honest perception of what is really the case. (Murdoch 1970: 38)

Murdoch’s suggestion here is that ‘impersonal quasi-scientific knowl-

edge’ is not all there is to knowledge. Indeed, she claims that the knowl-
edge that is morally significant (that to which ‘goodness is connected’)
and that reveals ‘what is really the case’ is not such value-neutral knowl-
edge at all. In saying this, she rejects the idea that something like Williams’
absolute conception will reveal all truths that are available to be known.
Moral knowledge, that is, might require concepts that make sense only
within the moral life.
Murdoch parodies the idea that all morally significant facts will be
revealed by value-neutral perception by suggesting that it models morality
on something like a simple shopping trip:

On this view one might say that morality is assimilated to a visit to a shop. I
enter the shop in a condition of totally responsible freedom, I objectively
estimate the features of the goods, and I choose. The greater my objectivity
and discrimination the larger the number of products from which I can
select….I find the image of man which I have sketched above both alien and
implausible. (Murdoch 1970: 8–9)

Again, Murdoch’s suggestion here is that value-neutral perception will

not reveal all that is of moral significance. On her account, the moral life
is not reducible to a set of choices made between the same discrete and
neutrally evaluable objects in the way that shopping might be. Rather, the
very objects and features one picks out are morally significant, and

12
Mulhall (2000) argues that, for Murdoch, we are continuously engaged with moral
value, and that this is a core tenet in her rejection of the existence of a distinction between
fact and value.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 49

sustained loving attention may reveal very different objects to those one
initially seemed to see.
On Murdoch’s account, many objectively real objects and qualities can
be understood only from within the perspective of a human, value-laden
conceptual scheme. Broackes (2012) describes this as Murdoch asserting
that “we should allow the world to contain all that meets the gaze of a just
and loving moral perceiver” (Broackes 2012: 47); according to Murdoch,
the objectively real includes that which is perceived from a personal per-
spective. Murdoch’s claim is that a human moral scheme is necessary to
fully perceive reality. For example, the qualities that M comes to see in D
are real qualities that D possesses, but could not be assimilated into an
impersonal or unloving account of D. There is no more basic description
that might capture what it means for D to be ‘delightfully youthful’ and
certainly no non-evaluative equivalent.13 For Murdoch, the applicability of
moral concepts cannot be understood from outside the moral schema
itself. The very concepts necessary for understanding the world themselves
can themselves be understood only ‘in depth’, from the perspective of an
agent embedded in moral practices who is to some extent virtuous.14

3.5   Love as a Virtue and a Perceptual Sensitivity

Murdoch claims, then, that love has an irreducible epistemic role: it
involves knowledge or perception of reality. This reality is to be under-
stood as existing independently of being loved, but perceptible only to the
person who lovingly attends to it. I have suggested that neither Velleman’s
nor Hopwood’s interpretations of Murdochian love do justice to both the
broad scope of love’s epistemic role in her thinking and to her realism
about what is to be perceived. In this section, I will suggest that Murdoch
thinks of love as a virtue and outline the conception of virtue that she has
in mind. On her account, the virtues in general are epistemic and hierar-
chically ordered traits, and love occupies a particularly lofty position in the
hierarchy of the virtues.

13
Murdoch thus claims that moral philosophers’ task is “the provision of rich and fertile
conceptual schemes” (Murdoch 1970: 45): had M possessed only concepts such as ‘juvenile’
and ‘vulgar’, she would have been unable to recognize that D is in fact ‘refreshingly simple’
and ‘gay’.
14
Murdoch’s view is similar to that defended by McDowell (1979, 2011). In “Aesthetic
Value, Objectivity and the Fabric of the World” (1998), McDowell explicitly rejects Williams’
absolute conception of objectivity.
50 C. MASON

A seldom remarked-upon but significant feature of Murdoch’s concep-

tion of love is that she repeatedly refers to love as a virtue and includes it
among lists of the virtues in discussions of the moral life. This, I want to
suggest, is central to understanding how Murdoch retains her commit-
ment to moral realism while claiming that love involves knowledge. As I
have discussed, Murdoch considers love to be a desirable quality. Further,
she suggests that it is not merely desirable or pleasant, but that it is morally
necessary: it is ‘commanded’. As such, she thinks of it as a virtue:

All just vision, even in the strictest problems of the intellect, and a fortiori
when suffering or wickedness have to be perceived, is a moral matter. The
same virtues, in the end the same virtue (love), are required throughout,
and fantasy (self) can prevent us from seeing a blade of grass just as it can
prevent us from seeing another person. (Murdoch 1970: 70)

As we deepen our notions of the virtues we introduce relationship and hier-

archy. Courage, which seemed at first to be something on its own, a sort of
specialised daring of the spirit, is now seen to be a particular operation of
wisdom and love….It would be impossible to have only one virtue unless it
were a very trivial one such as thrift. (Murdoch 1970: 95)

Here, Murdoch refers to love as a virtue and lists it among more com-
monly recognized virtues such as courage and wisdom.15
On this conception of love, it is not simply an episodic attitude, but a
deeply important character trait. The virtues are often thought of as traits
that involve certain dispositions: dispositions to think, act, perceive, and
feel in certain ways. In the first aforementioned quotation, Murdoch sug-
gests that love is required for ‘just vision’. Extrapolating from this and
from cases such as M and D, we come to a conception of the kind of virtue
that Murdoch has in mind. Murdoch conceives of love as a virtue that
entails the disposition to know, grasp, or understand an object of attention
ever more adequately. On the Murdochian account, then, love involves a
disposition to see truly, to (progressively) perceive individuals as they
15
I am here leaving open exactly what kind of virtue ethicist Murdoch is, as well as the role
of the virtues in her overarching account of ethics. My argument depends only on the uncon-
troversial ideas that she is deeply impressed by the importance of the virtues and that she
regards them as having a crucial role in the moral life. McLean (2000) offers an argument
against identifying Murdoch as a virtue ethicist, noting that she is more influenced by Plato
than by Aristotle and is therefore at odds with neo-Aristotelian virtue ethics. This, however,
is no reason to think that she is not some form of virtue ethicist.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 51

really are. The connection between love and knowledge is thus intimate:
love is necessarily truthful because love is (at least partly) constituted by
progression toward ever more adequate knowledge of its object.
We can shed light on the connection between love and moral knowl-
edge by considering the general role of the virtues in Murdoch’s thinking.
For Murdoch, the virtues are reliable sensitivities to certain features of the
world, and as a virtue, love involves such a perceptual sensitivity. On this
account, the virtues in general therefore look as much like epistemic dis-
positions as affective or motivational dispositions. Murdoch states: “virtue
is the attempt to pierce the veil of selfish consciousness and join the world
as it really is” (Murdoch 1970: 93). Elsewhere she writes that “anything
which alters consciousness in the direction of unselfishness, objectivity and
realism is to be connected with virtue” (Murdoch 1970: 84). For Murdoch,
virtues are thus highly epistemically significant traits: they are traits that
enable a kind of perception that arises only from a human and normatively
rich standpoint. Given this conception of virtue, love is necessarily truth-­
conducive: one can only perceive or be sensitive to real features of
the world.16
This knowledge or perception is connected with action; the perceptual
sensitivities that constitute virtues are manifested in dispositions to act.
The person who perceives the true extent of these demands, Murdoch
suggests, will act in ‘obedience’ to reality:

[T]he idea of a patient, loving regard, directed upon a person, a thing, a

situation, presents the will not as unimpeded movement but as something
very much more like ‘obedience’. (Murdoch 1970: 40)

In other words, Murdoch suggests that the agent who perceives the full
moral significance of a situation is often not left with an open choice about
how to respond. In order to discern the true moral contours of a situation
in the first place, the agent must attend in a way that is loving. Attending
lovingly is motivationally and affectively laden; it eventuates in ‘obedience’
to the moral demands of what is perceived.17

16
A similar conception of virtue is proposed by McDowell (1979), who also understands
the virtues as perceptual sensitivities and argues for the claim that ‘virtue is knowledge’.
17
It seems plausible that there will be degrees of love, so not all love will entail complete
moral motivation. But insofar as one is loving, one will be motivated to act in accordance
with what is perceived.
52 C. MASON

Murdochian love is not, however, simply a form of clinical precision. In

relation to the example of M and D, Murdoch notes that “what M is ex
hypothesi attempting to do is not just to see D accurately but to see her
justly or lovingly” (Murdoch 1970: 23). The ‘just’ in this quote is signifi-
cant: seeing lovingly is not opposed to accuracy, but is a far richer form of
vision, a form of vision reaching beyond simple accuracy, which may be
difficult to achieve. In approaching the object of love from the loving
perspective, one approaches it from a position into which is built a com-
mitment to understanding the object justly and in its full complexity.
Moreover, the affective richness of the loving perspective allows for a
depth in one’s grasp of one’s concepts that is transformative of them. The
loving perceiver does not take the same concepts that could be grasped
from a detached perspective and apply them to a fixed scenario, but they
rather have conceptual resources that differ from those of the unloving
perceiver, which transform what one can perceive.18
To illustrate this, compare the ballet lover’s experience while watching
Swan Lake to the experience of a reluctant audience member with no
interest in ballet. While both would have a visual experience of the same
thing, they would see very different things. The ballet lover might see
graceful arabesques and lively grand jetés within an innovative production
of the ballet, whereas the disinterested audience member might simply see
dancers moving and jumping. Even if we imagine that the disinterested
audience member has better eyesight, so that their perception of the ballet
is more visually accurate and they are able to track more precisely the exact
movements that the dancers make, we would plausibly think that their
vision and understanding of it was importantly lacking. The loving atten-
tion that the ballet lover pays to the performance plausibly alters what they
discern in it, enabling them to better understand the ballet as a whole.
To love, for Murdoch, therefore entails attending to particular objects
from a virtuous perspective which involves an affective component. This
affective component includes generosity and an appreciative understand-
ing of the object of love. By viewing the object from this perspective, the
good qualities that it genuinely possesses become visible. Viewing others
in this way enables one to perceive real qualities that they genuinely

18
Murdoch writes, “Knowledge of a value concept is to be understood…in depth, and not
in terms of switching on to some given impersonal network….We do not simply, through
being rational and knowing ordinary language, ‘know’ the meaning of all necessary moral
words” (Murdoch 1970: 29).
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 53

possess, but which, without attending lovingly, one will not be sensitive
to. For example, in Othello, were one to view Othello from a detached,
impersonal standpoint, his character would undoubtedly be unappealing.
However, the play derives its power and its tragedy from enabling one to
perceive him from a loving perspective, from which he can be seen as par-
tially noble, yet at the same time deeply mistaken and cruelly blind. These
qualities are a genuine part of the object of perception, but they are not
visible from a perspective external to love.
This account of love as a virtue that is a reliable perceptual sensitivity
enables Murdoch to maintain that love is epistemically beneficial. However,
this on its own does not entail that love is necessary for true vision, nor that
it is uniquely epistemically significant. In the following sections, I shall
suggest that these features of love can be understood as a result of
Murdoch’s acceptance of the unity and hierarchy of the virtues, respectively.

The Unity of the Virtues

Many virtue ethicists have been tempted by the thought that the virtues
are somehow unified.19 Murdoch too understands the perceptual sensitivi-
ties constituting the virtues in this way. In discussing this, she claims that
an examination of everyday moral virtues reveals ways in which they are
deeply intertwined and ordered. Again, this is suggested by the above-­
mentioned text:

As we deepen our notions of the virtues we introduce relationship and hier-

archy.…It would be impossible to have only one virtue unless it were a very
trivial one such as thrift. (Murdoch 1970: 95)

Murdoch’s claim is thus that no single virtue can be understood, let

alone possessed, in isolation.20 On her conception, the virtues are percep-
tual sensitivities to certain features of the world. But the features that call
for kindness, for example, must be understood in relation to those that call

19
For more recent defenses of the unity of virtue, see Badhwar (1996), Wolf (2007), and
Toner (2014). Badhwar and Wolf defend qualified versions of the thesis. For skepticism
about the unity of the virtues, see Sreenivasan (2009).
20
She adds the caveat “unless it were a very trivial one such as thrift” (Murdoch 1970: 95).
However, the kind of thrift that is virtuous plausibly involves other virtues such as prudence
and a proper appreciation of goods (in order to distinguish appreciative thrift from mere
stinginess).
54 C. MASON

for justice, and so on: the fullest form of kindness will be sensitive to the
demands of justice.
Murdoch argues not only that the virtues cannot be defined in isolation
but that they cannot be possessed in isolation: one cannot be truly coura-
geous, for example, without also having the wisdom to know how and
when to act courageously. This does not entail that one cannot possess any
virtue to any degree without possessing the other virtues with which they
are conceptually interconnected, but that the virtues cannot be possessed
in isolation insofar as one could not fully possess any virtue without pos-
sessing the virtues with which it is interconnected. She writes, for instance:

[T]he best kind of courage (that which would make a man act unselfishly in
a concentration camp) is steadfast, calm, temperate, intelligent, loving.
(Murdoch 1970: 57)

On this view, lack of one virtue can impose a limitation on the extent to
which one can possess another, and the fullest form of any virtue will
involve further virtues.
For Murdoch, love, as a virtue, is therefore interrelated with every
other virtue: to be loving, in the fullest and truest sense, involves being
just, wise, honest, and so forth. Love, on this account, is therefore neces-
sary for the full possession of any virtue. This yields a sense in which love
is always epistemically required: it is a perceptual sensitivity, and full pos-
session of the perceptual sensitivities that are the other virtues also requires
love. Love is not therefore required only on odd occasions in order to
perceive a narrow set of features of the world, but it is necessary for all
fully virtuous perception. Insofar as the virtues are unified, love allows one
to perceive the world justly, courageously, and compassionately, and is
therefore epistemically valuable in enabling all of these sensitivities.

The Unique Significance of Love

Love, in Murdoch’s view, is thus deeply intertwined with all other virtues.
However, for Murdoch, love occupies a unique position among the vir-
tues: it is love in particular that is identified as “a central concept in mor-
als” (Murdoch 1970: 2). The thesis of the unity of the virtues alone does
not provide reason to set love apart from any other virtue. It suggests that
love is bound up with every other virtue but that the same is true of all
virtues, since they are all interconnected. Nonetheless, in the aforemen-
tioned quotation Murdoch claims that deepening our concepts of the
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 55

virtues introduces not only relationship between the virtues but also ‘hier-
archy’. ‘Hierarchy’ suggests that some virtues are more fundamental than
others and play a more significant unifying role within the realm of the
virtues. In Murdoch’s scheme, love occupies a special position in this hier-
archy: love is the form of all the virtues and has a particularly close connec-
tion to the Good itself.
Murdoch’s suggestion is that love occupies a special position in the
hierarchy of the virtues because the formal object of love is simply the real.
On her account, love is a form of perception whose object is the real—that
which exists outside of oneself and constrains one’s will: “love…is the
discovery of reality” (Murdoch 1959: 52). Elsewhere, she discusses “the
real which is the proper object of love” (Murdoch 1970: 68). As such, all
virtues are forms of love, for all virtues involve perception of the real. All
perception is perception of the real, and therefore all virtues are forms of
love. One can attend lovingly to any object in the world, and for any
object, loving attention will be morally and epistemically appropriate,
allowing one to see it as it truly is and thus respond in a suitable way.
According to Murdoch, the form or method of all the virtues is love:
loving attention is necessary for all true vision. That is, all virtues require
and involve true vision of the world, and for Murdoch, true vision means
that they involve loving attention. On her account, love is therefore a nec-
essary component of any virtue since it is the truthful vision that allows
perception of the particular features of the world sensitivity to which con-
stitutes the particular virtues. Since love is necessary for and an integral
constituent of every other virtue, love has a special place within the hier-
archy of the virtues, involving a unique contribution to all virtues.
Murdoch thus states: “‘Good’: ‘Real’: ‘Love’. These words are closely
connected” (Murdoch 1970: 42). On her account, love is a form of atten-
tion to and perception of the real, and the good is to be found in the deep
configurations of the real. Love, for Murdoch, is a form of attention to
particulars, and as such, it is the method of all the virtues. According to
Murdoch, to be loving is to attend virtuously to the real, and loving atten-
tion to the real reveals entities that make moral demands on the perceiv-
er.21 Murdoch’s justification of love’s epistemic and moral significance is
therefore dependent on her account of love as the form of all virtues.

21
Murdoch states, “Is there not nevertheless something about the conception of a refined
love which is practically identical with goodness? Will not ‘Act lovingly’ translate ‘Act per-
fectly’, whereas ‘Act rationally’ will not?” (Murdoch 1970: 102).
56 C. MASON

3.6   The Everyday Concept of Love: A Defense

of Murdoch

Murdoch thus presents an epistemically weighty account of love. She

understands it as involving perception of the real and conceives of the lov-
ing perceiver as progressing toward a deepened understanding of the
object. However, one might question whether what she is discussing is
genuinely love. Setiya (2013), for instance, raises the question of whether
Murdoch’s use of “love” is ‘quixotic’ and leaves this unanswered, and
Schauber explicitly claims that “Murdoch’s official, cognitive conception
of love is unfamiliar” (2001: 482). I shall address two particular skeptical
questions that might arise in this regard: firstly, can this account of love
allow for love’s affective dimension? Secondly, can this account explain
love’s selectivity, and particularly the problem that evil objects of attention
seem to pose? I shall suggest that Murdoch has the resources to respond
to each of these concerns.
Firstly, the affective dimension to love: one might worry that in under-
standing love as a perceptual sensitivity, Murdoch affirms its epistemic sig-
nificance at the expense of its affective role. However, love’s epistemic role
does not entail that it lacks an affective dimension. Indeed, Murdoch sug-
gests that an account on which cognition or perception is severed from
evaluation and affect is “both alien and implausible” (Murdoch 1970:
9).22 Identifying love with perception does not imply that it is not an emo-
tion, nor that it is not affectively significant; rather, perception itself, or the
knowledge thus gained, might be affective. Doring (2007) and de Sousa
(1987, 2002), for instance, argue that emotions are perceptions of value,
Roberts (1988) understands emotions as concern-based construals of
value, and Nussbaum (2001) argues that emotions are forms of evaluative
judgment. If emotions are understood on models such as these that unite
the epistemic and the affective, love can be both perceptual and emo-
tional. The phenomenal, affective, or ‘emotional’ character of love is not
therefore denied by understanding love as involving knowledge; conceiv-
ing of love as an emotion is compatible with thinking of it as being a kind
of perception.23

22
Indeed, Murdoch speaks of “obedience to reality as an exercise of love” (Murdoch 1970:
42), suggesting a close connection between love and action, a connection that plausibly goes
via the affective.
23
Murdoch is not alone in offering an epistemically laden account of love: Jollimore
(2011) also understands attending to the beloved in certain way as central to love.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 57

Murdoch’s conception of love as an epistemic state, then, is at least

compatible with the ordinary conception of love as an affective state. She
is, however, committed to the claim that one cannot truly love another
without being in the concomitant epistemic state, whereas it might seem
that the ordinary conception of love is only or primarily of an affective
state. Although Murdoch’s account is consistent with the ordinary con-
ception of love, one might therefore worry that it has a significantly differ-
ent emphasis.
In fact, however, we do ordinarily think of love as importantly involving
knowledge. It seems plausible that if someone were utterly unmotivated to
understand another, then however warm one’s feelings toward them, this
would fall far short of love: they would be failing to relate to the intended
object of love. Moreover, as Badhwar (2005) notes, “to the extent that
others are deceived about us we fail to be the actual objects of their love”
(Badhwar 2005: 60). Othello’s professed love for Iago, for instance, seems
to be based upon too pervasive a misunderstanding of who Iago is to truly
love him. Othello’s profoundly mistaken beliefs about Iago prevent him
from knowing Iago and thus form a barrier preventing him from loving
Iago. It is thus plausible that love at least has the aim of knowing or
understanding the other, and a love that did not involve any insights about
the beloved would seem questionable.
Further, although the everyday conception of love is closely related to
an affective state, there is reason to think that it is not reducible to this.
Naar (2013), for instance, argues that considerations such as the historical
nature of love, its ability to permeate one’s identity, and its persistence
across both time and temporal disruptions suggest that it is not merely an
occurrent affective state. As he notes, love is not a state that one could be
in for only five minutes and is conceived of as persisting throughout dis-
ruptions such as depression or doubt. The ordinary concept of love is
therefore not reducible to its affective dimension.
Moreover, some significant elements of paradigmatic instances of inter-
personal love are straightforwardly epistemic. As a friend or lover, one
discerns features in another beyond those which would be available to an
unloving observer, revealing a deeper knowledge of who the person is.
Some prominent non-philosophical descriptions of love focus on this epis-
temic dimension to love. Take, for example, Jane Austen’s description of
Darcy’s growing love for Elizabeth in Pride and Prejudice:
58 C. MASON

Mr Darcy had at first scarcely allowed her to be pretty; he had looked at her
without admiration at the ball; and when they next met, he looked at her
only to criticise. But no sooner had he made it clear to himself and his
friends that she had hardly a good feature in her face, than he began to find
it was rendered uncommonly intelligent by the beautiful expression of her
dark eyes…he was forced to acknowledge her figure to be light and pleasing;
and in spite of his asserting that her manners were not those of the fashion-
able world, he was caught by their easy playfulness.

In this passage, Austen reveals Darcy’s emerging love for Elizabeth by

describing his increasing disposition to perceive her good qualities. The
everyday conception of love does not therefore seem to identify it solely as
an affective state, but to involve perception.
Secondly, a critic might claim that the ordinary conception of love
involves selectivity: we do not love all equally, and we feel justified in limit-
ing loving relationships to particular people. For Murdoch, on the other
hand, love is morally ‘commanded’ for every object of attention. She states
that the virtuous agent, like the artist, sees their objects lovingly “whether
they are sad, absurd, repulsive or even evil” (Murdoch 1970: 66).
However, lovingly attending to all of reality does not imply that one must
always express love in the same way, or that love needs to always take the
same form. Loving one’s friends and one’s children, for example, will
involve very different relationships. It seems plausible that the everyday
conception of love covers a variety of phenomena, and that romantic and
sexual love, for example, involve far more than the basic moral case. In
these cases, the aspects of love that Murdoch identifies might be necessary,
but not sufficient, conditions for love. Thus, that selectivity is part of the
everyday conception of romantic love need not be in tension with loving
attention being ‘commanded’ for every object of attention.
Moreover, love’s selectivity is in part explained by the differing relation-
ships that we have with others. The beloved’s attitude toward the lover
plausibly affects the agent’s capacity to lovingly perceive them. The
beloved’s behavior and attitudes can plausibly enable or make difficult lov-
ing perception of them. Certain ways of acting open one up to others,
express one’s identity, and encourage engagement, while other ways of
acting (indifference, taciturnity, aloofness) discourage the perceiver from
attending lovingly. Although love is morally and epistemically necessary,
there is therefore an explanation for the selective way most people love:
the way another person acts can assist or hinder the lover in lovingly
attending to them.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 59

Nonetheless, an objector might urge that this does not account for evil
objects of attention, objects that seem unworthy of love. The idea that
such evil objects morally and epistemically ought to be lovingly perceived
seems to be far less obviously attractive than the idea that one’s friends and
partners ought to be lovingly perceived; such objects do not seem to be
lovable.24
One response to this is that as well as identifying love with knowledge
of the real, Murdoch seems to identify ‘the real’ with ‘the Good’: “‘Good’:
‘Real’: ‘Love’. These words are closely connected” (Murdoch 1970: 42).
Underlying Murdoch’s work runs a deep optimism in the reality and mag-
netic power of ‘the Good’, which might justify the idea that loving atten-
tion reveals objects that are ultimately worthy of love. However, I shall set
aside this option, since it involves theoretical commitments which many
would be hesitant to accept, and instead focus upon whether, if the real
and the good are extricable, one might still conceive of love as knowledge
of the real.
Crucially, this objection depends upon an un-Murdochian model of
love. Understanding love as a reliable sensitivity to the real does not entail
that one must find the object of one’s love to be ‘lovable’. For instance, in
the M and D example, Murdoch allows that attending lovingly to D might
lead M to conclude that her daughter-in-law is indeed unworthy. In the
same way that virtues such as justice might require negative appraisals and
emotions, so too a properly loving response might include ultimately neg-
ative evaluations.25 Attending lovingly does not entail that one will ulti-
mately conclude with a positive appraisal of the object of attention, but
that the genuinely positive features of the object that are there to be seen
will be increasingly fully perceived: the ultimate appraisal of the object will
be just and truthful—but not necessarily positive. Indeed, the connection
to virtues such as honesty and justice suggests that loving necessarily
involves possible negative evaluations as well as positive ones. However,
these will be situated within a vision of the other that does justice to the
complex whole. It does not seem implausible that it is right to perceive
even things that are overall unpleasant or evil in this way.

24
Chappell (2018) takes such objections to give reason to think that Murdoch does not,
after all, identify love with knowledge or take love to be necessary for knowledge.
25
Wolf (2015) notes that to love, and to lovingly attend to another, need not entail finding
them wholly lovable or admirable. Indeed, she suggests that the best kind of love involves a
clear-eyed awareness of the beloved’s flaws.
60 C. MASON

Finally, the idea that no one is an inappropriate object of love is far from
peculiar to Murdoch. Perhaps the most famous ethical advice in the
Gospels is found in Jesus’ injunction: “But I say to you, Love your ene-
mies and pray for those who persecute you” (Matthew 5:44, NRSV),
which is surely a case of morally commanded love for evil objects of atten-
tion.26 In this context, it seems that Murdoch’s account of love coheres
with features of the familiar concept of it. If it is embedded in an ordinary
conception of love that love for one’s enemies is possible, then Murdoch’s
account seems like a natural development of the everyday conception of
love, and a development that may shed new light upon it.

3.7   Conclusion
I have argued that Murdoch’s claims about love’s epistemic role can thus
be understood in relation to her conception of virtue. On her account,
love is a virtue, and as such involves a perceptual sensitivity to objective
features of reality. Moreover, Murdoch conceives of the virtues as unified,
and of love as occupying a special position in the hierarchy of the virtues,
which explains her contention that love is of unique moral and epistemic
significance. However, Murdoch does not suggest that virtues attune one
to features of reality that could be discerned by any neutral or impersonal
perceiver; for her, there are objective features of reality that will be percep-
tible only from within a human moral schema. The loving agent’s concep-
tual resources themselves are transformed by loving attention. The
apparent tension between love’s epistemic role and objectivity is thus
resolved, since on Murdoch’s account love is personal but nonetheless
involves an openness to the real. Although this account of love can seem
surprising, it is nonetheless a rich and interesting account that is consistent
with core components of the everyday conception of love.

References
Badhwar, Neera K. 1996. The Limited Unity of Virtue. Noûs 30: 306–329.
———. 2005. Love. In The Oxford Handbook of Practical Ethics, ed. Hugh
LaFollette, 42–69. Oxford: Oxford University Press.

26
This congruence between Murdochian love and Christian love is unsurprising given that
Murdoch’s conception of loving attention was influenced by Simone Weil, a deeply religious
thinker.
 3 IRIS MURDOCH AND THE EPISTEMIC SIGNIFICANCE OF LOVE 61

Bagnoli, Carla. 2003. Respect and Loving Attention. Canadian Journal of

Philosophy 33: 483–515.
Broackes, Justin. 2012. Iris Murdoch, Philosopher. Oxford: Oxford University Press.
Chappell, Sophie Grace. 2018. Love and Knowledge in Murdoch. In Murdoch on
Truth and Love, ed. Gary Browning, 89–108. Basingstoke: Palgrave Macmillan.
Clarke, Bridget. 2012. Iris Murdoch and the Prospects for Critical Moral
Perception. In Iris Murdoch, Philosopher, ed. Justin Broackes. Oxford: Oxford
University Press.
De Sousa, Ronald. 1987. The Rationality of Emotion. Cambridge, MA; London:
MIT Press.
———. 2002. Emotional Truth. Proceedings of the Aristotelean Society 76: 247–263.
Doring, Sabine. 2007. Seeing What to Do: Affective Perception and Rational
Motivation. Dialectica 61: 363–394.
Hopwood, Mark. 2014. Love’s Work: Eros and Moral Agency. ProQuest
Dissertations and Theses.
———. 2017. “The Extremely Difficult Realisation That Something Other Than
Oneself Is Real”: Iris Murdoch on Love and Moral Agency. European Journal
of Philosophy 26: 477–501.
Jollimore, Troy. 2011. Love’s Vision. Princeton: Princeton University Press.
Jordan, Jessy. 2014. Reconsidering Iris Murdoch’s Moral Realism. Journal of
Value Inquiry 48: 371–385.
Keller, Simon. 2004. Friendship and Belief. Philosophical Papers 33: 329–351.
McDowell, John. 1979. Virtue and Reason. The Monist 62: 331–350.
———. 1998. Mind and World. Cambridge, MA; London: Harvard
University Press.
———. 2011. Colors as Secondary Qualities. In The Possibility of Philosophical
Understanding, ed. Wai-Hung Wong, Jason Bridges, and Niko Kolodny,
217–231. Oxford: Oxford University Press.
McLean, Mark. 2000. On Muffling Murdoch. Ratio 13: 191–198.
Mulhall, Stephen. 2000. Misplacing Freedom, Displacing the Imagination: Cavell
and Murdoch on the Fact/Value Distinction. Royal Institute of Philosophy
Supplement 47: 255–277.
Murdoch, Iris. 1956. Vision and Choice in Morality. Proceedings of the Aristotelian
Society 30: 32–46.
———. 1959. The Sublime and the Good. Chicago Review 13: 42–55.
———. 1970. The Sovereignty of Good. London: Routledge & K. Paul.
———. 1992. Metaphysics as a Guide to Morals. London: Chatto & Windus.
Naar, Hichem. 2013. A Dispositional Theory of Love. Pacific Philosophical
Quarterly 94: 342–357.
Nussbaum, Martha. 2001. Upheavals of Thought: The Intelligence of the Emotions.
Cambridge: Cambridge University Press.
62 C. MASON

Roberts, Robert C. 1988. What an emotion is: A sketch. Philosophical Review

97: 183–209.
Schauber, Nancy. 2001. Murdoch’s Morality: Vision, Will, and Rules. The Journal
of Value Inquiry 35: 477–491.
Setiya, Kieran. 2013. Murdoch on the Sovereignty of Good. Philosophers
Imprint 13: 1–21.
Sreenivasan, Gopal. 2009. Disunity of Virtue. The Journal of Ethics 13: 195–212.
Stroud, Sarah. 2006. Epistemic Partiality in Friendship. Ethics 116: 498–524.
Toner, Christopher. 2014. The Full Unity of the Virtues. The Journal of Ethics
18: 207–227.
Velleman, James David. 1999. Love as a Moral Emotion. Ethics 109: 338–374.
Weil, Simone. 1956. The Notebooks of Simone Weil, trans. Arthur Wills. New
York: G. P. Putnam’s Sons.
Williams, Bernard. 1978. Descartes: The Project of Pure Enquiry.
Harmondsworth: Penguin.
———. 1981. Persons, Character and Morality. In Moral Luck: Philosophical
Papers 1973–1980, 1–19. Cambridge: Cambridge University Press.
Wolf, Susan. 2007. Moral Psychology and the Unity of the Virtues. Ratio
20: 145–167.
———. 2015. The Variety of Values. Oxford: Oxford University Press.
 CHAPTER 4

‘Love’ as a Practice: Looking at Real People

Lotte Spreeuwenberg

4.1   An Ameliorative Project

Recent philosophical discussions about love often focus on reasons to love
a particular person. Some philosophers argue that we do not have reasons
to love (Frankfurt 2009; Smuts 2013; Zangwill 2013), but rather that our
love for that particular person gives us reasons. Harry Frankfurt argues
that what we love is important to us just because we love it. Others argue
that we do have reasons to love the particular people we love, but disagree
on what these reasons consist in. For example, the reason for love is the
properties of the object of our love (Abramson and Leite 2011; Jollimore
2011; Keller 2000), such as being funny, or having beautiful eyes, or our
relationship with this person (Kolodny 2003).
In these recent discussions philosophers of love seem to primarily focus
on ‘reasons to love person X’. But what about ‘reasons to love, period’?
Focusing on why we love and what love is in general, enables us to look at
what we find meaningful in love, instead of what we find valuable in the
beloved. Focusing on the lover instead of the beloved could improve the
way we love in general, improving our relationships with all those

L. Spreeuwenberg (*)
Centre for Ethics, University of Antwerp, Antwerp, Belgium
e-mail: Lotte.Spreeuwenberg@uantwerpen.be

© The Author(s), under exclusive license to Springer Nature 63

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_4
64 L. SPREEUWENBERG

particular X’s. Focusing on the loving agent could help answer questions
such as ‘what does it mean to love?’ and ‘how can we love better?’.
Apart from describing what love is, one could independently argue
about what kind of love would be more or less valuable. We could catego-
rize better and worse forms of love. In this chapter this normative dimen-
sion of the concept ‘love’ is considered. Such a project is not trying to
formulate the description of ‘love’, but it is focusing on what would be
better or worse forms of loving. Furthermore, this chapter is what Sally
Haslanger has called an ‘ameliorative’ project (Haslanger 2012). Such a
project involves trying to formulate a concept that best suits the point of
having such a term. What is the purpose of talking about love? An amelio-
rative project requires actively making decisions about what to mean when
using it. How can we change the world around us for the better and
improve how we use the concept ‘love’? In this sense, ameliorative proj-
ects can be important for social progress. What use of the word ‘love’
could improve the way we love, and how could it impact society?
‘All about love’ by bell hooks is such an ameliorative project about love.
Using personal anecdotes and psychological and philosophical ideas, she
criticizes the way in which ‘love’ is used in today’s society—which is,
according to her, ‘without much meaning’, for example, when referring to
how much we like our favorite food, color, or sports—and instead argues
that if we all came to the agreement that ‘love’ is a verb rather than a noun,
then we would all be happier (hooks 2001: 4). hooks believes love is more
of an interactive process and clarifies why society needs to adopt this use
of the word love.
What should this particular ‘verb’ consist in? In line with hooks I argue
that we would all love better if we think of love as a verb: love as an activity
of attending to one another. Love as an ongoing practice, a process. I turn
to two famous contemporary analytic philosophers, who have argued for
different but related accounts of love. By comparing David Velleman’s and
Iris Murdoch’s account of love I argue that Velleman’s account is not suit-
able for the ameliorative project, while Murdoch’s account enables us to
be better lovers. I argue that better love consists of an activity of loving,
instead of a passive evaluation. While love can be understood in many
ways, at least one aspect of it is captured in the slogan: ‘love’ is a verb. This
slogan captures the idea that loving is an activity and furthermore a spe-
cific activity: loving means engaging in an ongoing practice of loving atten-
tion, a process that requires continuous work. I will argue that Murdochian
love is not only valuable for philosophers or people who are concerned
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 65

with being morally good—which is Murdoch’s focus—but also particu-

larly valuable to ordinary lovers, to people who want meaningful loving
relationships.

4.2   Love, Value, and Looking

In ‘Love as a Moral Emotion’, Velleman attempts to assemble elements of
both Murdoch and Kant into an account of love as a moral emotion
(Velleman 1999). This famous contemporary account of love is meant to
address and combat Frankfurt’s position: love is not a response to reasons
but is in fact the basis of all reasons. On Frankfurt’s account, the lover
cares for the beloved, desires their well-being, and, in doing so, comes to
confer value upon the beloved. Against this bestowal view of love, Velleman
argues for the appraisal view of love inspired by Iris Murdoch. Murdoch,
Kant, and Velleman all allow that value may be discerned or figuratively
seen, as Tony Milligan has observed in his analysis of Velleman’s use of
Murdoch’s account of love (2013: 113). Velleman adopts Murdoch’s idea
that this value may be seen by ‘really looking’ (Velleman 1999: 343).
Murdoch’s account of love and Kant’s account of respect are taken by
Velleman to be complementary ways in which we recognize the inherent
value of persons. Velleman takes different features from both philoso-
phers’ work and is not alone in placing Murdoch’s ideas within a Kantian
perspective.1
Velleman, being a Kantian, argues that our value, our dignity as per-
sons, consists in our rational nature. One important way in which we exer-
cise our rational nature is to respond with respect to the dignity of other
persons. We are aware of the incomparable value in a person as a rational
being, and this awareness arrests our motives of self-interest and thereby
prevents us from treating him as a means to our ends. This is what Velleman
(and Kantians in general) call respect. Velleman argues that love is similarly
a response to the dignity of persons, and as such it is the dignity of the
object that justifies love. Velleman suggests that love, like respect, is an
appreciation for the capacity to be “actuated by reasons” (p. 365), which
means a capacity for “appreciating the value of ends, including

1
Although Murdoch regarded herself a Platonist, many philosophers have argued that her
ideas are compatible with Kantian ideas or have used her ideas within Kantian perspectives
(cf. Bagnoli 2003; Grenberg 2014; Merritt 2017; and Milligan 2013).
66 L. SPREEUWENBERG

self-existent ends such as persons”. Velleman argues: “I find it plausible to

say that what we respond to, in loving people, is their capacity to love”
(p. 365).
For Velleman, love and respect are responses to the same value, but
they are different kinds of responses. Love arrests not our self-love, like
respect does, but rather our emotional self-protection. Love disarms our
emotional defenses, making us vulnerable, by responding to someone’s
dignity (p. 361). Velleman claims that:

Many of our defenses against being emotionally affected by another person

are ways of not seeing what is most affecting about him. This contrived
blindness to the other person is among the defenses that are lifted by love,
with the result that we really look at him, perhaps for the first time, and
respond emotionally in a way that’s indicative of having really seen him.
(1999: 361)

There is a clear parallel with Murdoch here. Velleman’s view of love as

arresting our self-protective egocentricity helps to explain why love is an
exercise in really looking, precisely as Murdoch claims. Velleman describes
someone having “stopped loving his wife” as having “stopped really look-
ing or listening” (p. 373). His account is clearly inspired by Murdoch,
who claims that to love is to redirect our attention outside ourselves, to
learn to perceive the truth about the world and see what there is outside
oneself (1971) and that love is an opening up in the sense that it is “the
extremely difficult realization that something other than oneself is real”
(1999: 215).
So Velleman’s account of love seems to be broadly Murdochian, but by
combining Murdoch’s and Kant’s theories he has worked himself into an
awkward position. His account of love is detached in a way that Murdoch’s
is not.2 It involves a certain abstraction of people that Murdoch specifically
avoids. While Murdoch argues that we should look at the unique and par-
ticular (1971), Velleman’s account tells us that what we love is essentially
a person’s dignity, something that all rational beings share. On Velleman’s
account, this is sufficient to speak of love.
The detachedness of Velleman’s account of love becomes especially
clear when he explains his notion of ‘selectivity’. Since for Velleman love
and respect are different rational responses to the same value, he should

2
Possibly Kant’s account of love is also not as detached as Velleman’s (Milligan 2013).
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 67

have an explanation for why the number of people we respect generally

outweighs the number of people we love. Velleman is fully aware of this
problem and argues that we cannot respond with love to the dignity of
every person we meet, nor are we required to. For Velleman, love is the
optional maximal response to others’ dignity, while respect is the required
minimum. Hence, he argues for the selectivity of love. With this concept
Velleman tries to account for personal, but not partial, love. Selectivity of
love means that a contingent fit takes place between the way some people
behaviorally express their dignity as persons and the way we happen to
respond to those expressions by becoming emotionally vulnerable to
them. The right sort of fit makes someone lovable by us (1999, p. 372),
and our responding with love is a matter of our ‘really seeing’ the other in
a way that we fail to do with others who do not fit us this way. It is impor-
tant to note that by ‘lovable’ Velleman does not mean ‘worthy of love’.
Instead, he means something like ‘able to be loved’.

Whether someone is lovable depends on how well his value as a person is

expressed or symbolized for us by his empirical persona. Someone’s persona
may not speak very clearly of his value as a person, or may not speak in ways
that are clear to us. (1999: 372)

Velleman’s point is that we have many reasons for being selective in

love, without having to find differences of worth among possible love
objects. The people whom we do not happen to love may be just as eligi-
ble for love as our own children, spouses, parents, and intimate friends, he
argues. “In merely respecting rather than loving these people, we do not
assess them as lower in value. Rather, we feel one emotion rather than
another in appreciation of their value. Loving some but not others entails
valuing them differently but not attributing different values to them, or
even comparing them at all” (1999: 372).
This explanation of love generates a few problems. Firstly, it seems to
be an unrealistic illustration of what love is, for it puts little responsibility
on the lover. It might be true that recognizing another’s dignity can take
some effort: we have to at least be willing to disarm our emotional defenses
in order for this to happen. However, let’s keep the ameliorative project in
mind. Taking love to be something so detached that we merely have to
recognize the dignity of a person, and so contingent that we have little
responsibility concerning whom we love and connect with, will not make
us better lovers—or so I will argue in the following sections. Other uses of
68 L. SPREEUWENBERG

the word ‘love’ are better suited to improve the way we love and to enable
societal change. We should play a bigger part in loving each other. Velleman
puts too much emphasis on both the contingent fit—and this being
dependent on the way other people behaviorally express themselves—and
the response to a value that all human beings share. We would be better
lovers when love is more than contingent: love requires actively looking,
engaging in an ongoing practice. Furthermore, loving goes beyond seeing
someone’s dignity; we are not finished with practicing love once we have
recognized a value that all human beings share.
Velleman’s piece has been extensively discussed, and other problems
have been addressed by, for example, Edward Harcourt (2009) or Elijah
Millgram (2004). I will shortly address a solution to one of these previ-
ously stated problems, because it unintentionally reveals the need for an
account of love as a more active engagement, seeing the particularities of
our beloveds. One problem is that Velleman cannot explain love’s personal
character: if love is justified by a property that all rational people share, it
seems to follow that it cannot matter which rational being one loves, and
this seems contrary to another expectation we have about love—that it
involves focus on the particularity of the loved individual (Lopez-Cantero,
manuscript). By reformulating Velleman’s account of love, Pilar Lopez-­
Cantero accounts for the personal character of love by (1) explaining
Velleman’s selectivity as narrative fit and (2) reformulating what is under-
stood as ‘rational nature’ and ‘empirical persona’. Although this reinter-
pretation describes love in a way that is, like Velleman’s, a contingent,
somewhat passive, emotion, it is useful for the ameliorative project here.
Lopez-Cantero’s reformulation of Velleman’s account could be used as a
clear example of why we need an account of love as an activity. She argues
that it is our narrative, and not, as Velleman argues, our empirical persona,
that fits with some and not with others. Lopez-Cantero suggests that the
incomparable value of another is directly perceived through their personal
narrative, which is a direct product of the agent: the narrator. The con-
cepts ‘narrator’ and ‘personal narrative’ play an equivalent role in Lopez-
Cantero’s theory to ‘rational nature’ and ‘empirical persona’ in Velleman’s
theory. Since the lover has her own narrative agency which aims at intel-
ligibility, some narratives will be particularly meaningful to her, and in that
case there will be a narrative fit with the beloved. In this reinterpretation,
what Velleman calls selectivity is explained by reference to an actual prom-
ise of meaning for the lover, instead of a contingent fit between the more
abstract ‘empirical persona’. If ‘empirical persona’ is reformulated in terms
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 69

of a narrative fit, the difference between love and respect on these accounts
also becomes clearer.3 Respecting people is valuing human dignity, some-
thing which every human being shares. According to Velleman, respect
and love share the same final objects, but Lopez-Cantero’s reformulation
of Velleman’s selectivity makes clear that this final object is only accessed
via something that is different in everyone—but is a direct product of that
final object. We value personal agency (as the equivalent of rational nature
in Velleman’s account), but love happens by evaluation of its product:
personal narrative. This evaluation will differ since every individual’s per-
sonal narrative is unique. Because love and respect are different evaluation
processes, it is possible to respect someone whom we do not love, argues
Lopez-Cantero.

4.3   Love as Passivity: Unrealistic Images

and Fantasies

How does ‘narrative fit’ show the need for love as an activity? I submit that
Velleman’s and Lopez-Cantero’s depictions of what it means to love are
problematic: according to them, if we do not have a fit with an individual,
it is because his or her empirical persona expresses their dignity poorly to
us. Their view is too passive on the side of the lover and ignores some
responsibility on the part of the loving agent. Lopez-Cantero’s account of
love as a narrative fit does not do away with this passive evaluation, because
our narratives on her account seem to be just contingently fitting.
However, the idea of fitting narratives enables us to paint a better picture
of what love is all about. Love is often nothing like a contingent fit,
whether that is between empirical persona or narratives. Without ruling
out that a contingent fit could be part of love, such a fit might not be
enough to be (or remain) a good lover. Love does not come easily, for
loving involves hard work!
Consider the love between Dante and Beatrice (famous for being
Dante’s muse). Dante was helplessly in love with Beatrice, but during his
life they only met a few times and only twice they had the shortest conver-
sation of greeting each other (Alighieri 1294). His deep love for Beatrice
became his reason to write poetry. More crucially, it became his reason to
be alive. In Dante’s poems, Beatrice appears before him as a ghostly

3
Which is considered a different problem with Velleman’s account, pointed out by Edward
Harcourt (2009).
70 L. SPREEUWENBERG

shadow, a half-goddess. She functions as a muse, watches over him and

guides him, gives him helpful instructions, or criticizes him. While Beatrice
is enormously valuable to Dante, his ideas of her merely consist in projec-
tion, prompted by his own desires and feelings. He closes his eyes to the
real Beatrice, while focusing on his fantasy of her. Could love not be more
valuable—especially if we were Beatrice—than Dante’s fantasy?
Unfortunately we cannot ask her, since her voice only can be heard in that
particular version Dante made of her: a half-goddess, merely functioning
for Dante’s sake.
However, we can ask ourselves and the people around us. Dante and
Beatrice’s love story is obviously a historic example, and it was a picture of
love painted a long time ago. But we would consider a relationship based
on unrealistic images of the beloved non-ideal today, too. In romantic
relationships, we are often blinded by being hopelessly infatuated, so that
numerous fantasies about our beloved surface, flooding more uncomfort-
able realities. I don’t want to claim that infatuation is not valuable at all,
but I think that most of us would deem a relationship based on unrealistic
fantasies, or these fantasies being the core of the relationship throughout
time, not the most valuable, meaningful kind of love for both (or more)
people involved. Surely the same goes for friendship: if we truly want to
connect, we better open ourselves to the reality of our friends. Poet and
feminist Adrienne Rich puts it beautifully when she writes:

An honorable human relationship—that is, one in which two people have

the right to use the word “love”—is a process, delicate, violent, often terri-
fying to both persons involved, a process of refining the truths they can tell
each other. It is important to do this because it breaks down human self-­
delusion and isolation. It is important to do this because in doing so we do
justice to our own complexity. It is important to do this because we can
count on so few people to go that hard way with us. (Rich 1995: 111)

In an honorable human relationship, two people have the right to use

the word love, writes Rich. And loving is not such an easy task to take up:
it is ‘the hard way’. And this does not solely apply to romantic or personal
love; it is applicable to societal issues as well. There are obstacles for look-
ing at the reality around us: physically, when we push particular groups or
people out of our sight (e.g. literally putting away refugees), and psycho-
logically, when we are focusing too much on ourselves, our needs and our
desires. When groups or human beings don’t understand, appreciate, or
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 71

accept each other, they are often keeping each other—or rather, one is
keeping the other—at a distance: we don’t really want to see the other. Or
putting it differently, we don’t really want to see each other’s reality,
because we are too comfortable with our own. Dante loves ‘the fantasy
Beatrice’—he does not love Beatrice. For the real Beatrice, Dante is not a
great lover, at all.
Suppose that Dante has recognized Beatrice’s ‘incomparable value as a
human being’. Furthermore, to Dante, their empirical personas are con-
tingently fitting. Dante doesn’t have to do anything for this love to
emerge: it just happens. He saw her and boom, love was in the air. Such a
feeling or happening has been described many times as love (just think of
any romantic comedy or pick any love song), but that doesn’t mean this is
the type of love or loving that is particularly meaningful to us. For Beatrice,
there is not much love to it. In Lopez-Cantero’s account, ‘empirical per-
sona’ makes way for ‘narrative fit’, but this is still a detached form of love.
The fact that their ‘fit’ happens contingently means that neither Dante nor
Beatrice had any part in it and bear no responsibility whatsoever. The big-
ger problem is that because of this lack of agency, Dante is not really
attending to Beatrice, his desires and needs shaping a self-serving fantasti-
cal image of her.
Velleman and Lopez-Cantero could argue that Dante’s love does not
count as love on their accounts: Dante is not really valuing Beatrice’s dig-
nity, he just thinks he is; or Dante and Beatrice’s narratives don’t really fit,
Dante just thinks they do. Dante thinks he’s in love, but on their account
he is not. But even if this were true, just think of the sort of love Velleman
does have in mind and whether this would make us better lovers. On
Velleman’s account responsibility on the part of the lover is missing. In
one of his examples he states: “I think that love naturally arises […]”
(1999: 361). Such a passive attitude in love is an obstacle for really looking
at each other. Will Dante ever see the real Beatrice?
Velleman’s account of love is descriptive, not normative, and here our
philosophical projects differ. However, even of descriptive projects we can
ask what they contribute to the concept in the real world. What do we
want ‘loving’ to mean? The ameliorative project requires actively making
decisions about what to mean when using ‘love’. How can we change the
world around us for the better and improve how we use the concept?
What use of the word ‘love’ could improve the way we love, and how
could it impact society? Velleman’s account of love does not suffice: not its
contingency, not its passivity. bell hooks is right when she states that while
72 L. SPREEUWENBERG

the word ‘love’ is most often defined as a noun, we would all love better if
we used it as a verb (2001: 4). Love is particularly meaningful to us when
we talk about it in terms of an attitude that one can take up: an ongoing
practice or process we can actively engage in. We better use ‘love’ as a verb
and it needs an active object.

4.4   Love as an Ongoing Practice: Steering Away

from the Ego

Let’s view the example of Dante and Beatrice from a Murdochian view of
love. For Murdoch, loving consists in looking beyond the ego, focusing
our attention on the particular and the unique. She holds that to love is to
redirect our attention, to learn to perceive the truth about the world, and
to see what there is outside one (1971). Constantly attending to our
needs, our desires, and our thoughts alters our perspective on what the
world is actually like and blinds us to the goods around us. Murdoch states
that “in the moral life the enemy is the fat relentless ego” (1971, p. 51)
and love, as focused attention, is steering away from the ego. We are often
so much focused on ourselves, our desires and needs, that we are blind to
the things and people around us. However, we do want to truly connect.
We appreciate it when the people around us are able to look beyond the
limits of their own world and see us for what we actually are. We want to
be truly seen, or at least we don’t want our lover’s needs or desires con-
stantly trumping our experiences, when we are in (any) relationship.
We should therefore think of love as actively attending: a process in
which we attend lovingly to our beloveds with an open gaze. This effort
does not necessarily require behavioral ‘proof’ that is visible to others.
Consider this famous passage from Iris Murdoch in The Sovereignty
of Good:

A mother (M) feels hostility to her daughter-in-law (D). M finds D quite a

good-hearted girl, but while not exactly common yet certainly unpolished
and lacking in dignity and refinement. D is inclined to be pert and familiar,
insufficiently ceremonious, brusque, sometimes positively rude, always tire-
somely juvenile. M does not like D’s accent or the way D dresses. M feels
that her son has married beneath him. […] Time passes, and it could be that
M settles down with a hardened sense of grievance and a fixed picture of D
[…]. However, the M of the example is an intelligent and well-intentioned
person, capable of self-criticism, capable of giving careful and just attention
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 73

to an object which confronts her. M tells herself: ‘[…] let me look again.’
Here I assume that M observes D or at least reflects deliberately about D,
until gradually her vision of D alters. […] the change is not in D’s behavior
but in M’s mind. D is discovered to be not vulgar but refreshingly simple,
not undignified but spontaneous, not noisy but gay, not tiresomely juvenile
but delightfully youthful, and so on. (1971: 16)

The example shows that M looks at D, she attends to D and focuses her
attention. She is trying to see D in a way that goes beyond her own projec-
tions, in a way that is not guided by her ego. It takes place in the inner life,
in M’s mind, but nevertheless is an action: she engages in the practice of
focusing her attention on D. It is unlikely that M would come to value D
in new ways unless she made the effort to look at her with an open gaze.
It is in this sense that love is a realization, an opening up in the sense that
it is “the extremely difficult realization that something other than oneself
is real” (1999: 215).
When we are focused on our own desires and needs, we fall prey to the
dangers of the ego: we make up fantasies in our minds. Love is meaningful
to us when we are able to steer away from our ego and perceive the par-
ticularity and uniqueness of a person, their reality. On Velleman’s account
of love, Dante is able to entertain self-serving fantasies of Beatrice. Love
here is something contingent, and it suffices to value dignity to speak of
love. However, Murdoch’s theory of love is a less detached version of the
concept, since we must adapt our concepts to the uniqueness of the par-
ticular people we meet. Dante should engage in the practice of loving
attention, looking at Beatrice with an open gaze, and consequently would
see more of the real Beatrice. What would have happened if Dante had
engaged in this practice? If he hadn’t been blinded by his own desires and
needs? If he had not let his ego guide him, but had actively looked at
Beatrice? Actively looked at her, opening up in the sense that he had ‘the
extremely difficult realization’ that Beatrice, someone other than himself,
someone outside his art, his emotions, and his intellect, was real. Our
desires and fantasies tend to make us blind to the things around us; make
everything around us fit the concepts that we already have or believe to be
true. But when we focus on the particular and the unique, we can come to
know new concepts and new realities, and it is more likely that we won’t
get stuck in our own self-serving worlds. Murdoch is trying to tell us that
it is not love that is blind, but our ego.
74 L. SPREEUWENBERG

4.5   Real People Versus Fantasies

Why is seeing each other’s reality so particularly valuable? Being fed up
with fantasies about certain persons or groups is a recurring theme in
fighting for equality. With such a fight for equality, often a call for atten-
tion comes along: look at us, hear us, see our (particular) truth. This call
for attention urges its addressees to attend to the reality of a certain group,
instead of projecting fantasies onto that group, fueled by the blindness
and egos of an oppressive group or society.
Some of the most important feminist works discuss the cultural myths
that have been around—or are still very much alive—about the reality of
women. In fact, the very notion of mysticism is in titles of some of the
most famous feminist works of the past century. With The Feminine
Mystique, Betty Friedan challenged the widely shared belief in the 1950s
that “fulfillment as a woman had only one definition for American women
after 1949—the housewife-mother” (1963). The phrase ‘feminine mys-
tique’ was created by Friedan to show the assumptions that women would
be fulfilled from their housework, marriage, sexual lives, and children. It
was said that women, who were actually feminine, should not have wanted
work, get an education, or have political opinions. By portraying that
image as mystique, Friedan showed the reality of many of these women,
namely, that they were dreadfully unhappy. In turn, Friedan was criticized
for being blind to the experiences of women other than those belonging
to the white middle class. bell hooks labeled Friedan’s project narcissistic
and self-indulgent and writes in Feminist Theory: From Margin to Center:
“She did not speak of the needs of women without men, without children,
without homes. She ignored the existence of all non-white women and
poor white women” (2000). Another book that specifically talks about the
harmfulness of blindness and fantasies is The Beauty Myth, in which Naomi
Wolf (1991) gives us a similar argument when describing societies’ fanta-
sies of female beauty. Although she wrote the book in the early nineties,
the argument never ceased to be relevant. The pressure that many women
feel to adhere to unrealistic social norms of physical beauty is still leading
to unhealthy behavior and an obsession with the female appearance for
both sexes, today. Many men (and women) have unrealistic beliefs about
the physical female beauty, the physical appearances of real human beings.
Fantasies, such as ‘women do not grow body hair’, ‘hair is smooth and
straight and certainly not Afro-textured’, or ‘all women are hourglass
shaped’, create detachment between real people. Firstly, fantasy obstructs
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 75

a connection because unrealistic beliefs make for unrealistic expectations

which could lead to disappointment. Secondly, it obstructs a connection
because many are trapped in the conflict between expectation and reality,
wanting to live up to expectations and thus constantly engage in self-­
masking or self-destructive behavior—in Wolf’s example: shaving, waxing,
lasering, dyeing, relaxing,4 a thirty-something-step beauty routine, and
dieting obsessively.
Now one might try to point out the irrelevance of my argument by
arguing that one does not want to love real human beings or that loving
fantasies is particularly valuable. Some might be particularly happy living
inside their fantasies, without harsh reality ruining self-serving dreams.
Furthermore, it takes some effort to actually look at others’ reality, espe-
cially from a privileged position in which the world is built for your com-
fort. Secretly, we might not want to learn new concepts. It is not my
intention to make the moral case that love should be love only between
real human beings and that it is, in any case, wrong to love fantasies. The
argument here is about better lovers and how we can improve how we use
the concept ‘love’. If we want love to be something meaningful between
real human beings, as opposed to self-serving fantasies, we need to take up
the activity of focusing our attention outside ourselves. Above all, privi-
leged persons have all the more reason to engage in an activity of loving
attention if they want to be better lovers. The problem often is, however,
that until privileged people (myself included in many respects, as a white,
educated, non-disabled person) are really looking, we are not aware that
we are so privileged, somewhat stuck in our own realities.
There is a positive argument in favor of the role of fantasies, too. Lisa
Bortolotti argues that optimistically biased beliefs can help us attain our
goals, based on literature on positive illusions in the perception of roman-
tic partners (2018). Bortolotti argues that optimistic beliefs lead to goal
attainment by sustaining our motivation to act after we experience set-
backs or when some of our cherished goals are under threat. Multiple
positive illusions could benefit our relationships in this way. One could
have optimism bias underestimating the likelihood of getting a divorce
even when we are well-informed about the high divorce rates in this soci-
ety (Baker and Emery 1993; Fowers et al. 2001). Such a bias may be sup-
ported by other positive illusions about the relationship: the relationship

4
Referring to the chemically straightening of tight curly (e.g. Afro-textured) hair, not to
relaxing as calming or unwinding activity.
76 L. SPREEUWENBERG

superiority bias occurs when we rate our relationship as better than most,
while we experience the love-is-blind illusion when we are blind to our
romantic partners’ faults and perceive our partners as better than average
in a number of domains, including intelligence and attractiveness (Buunk
and van den Eijnden 1997; Murray et al. 1996a, b; Rusbult et al. 2000).
We tend to idealize our partners’ qualities, and this could be beneficial for
the relationship we have with them, Bortolotti argues. The idealization of
romantic partners helps us continue to value the relationship as something
worth working on and is linked to more satisfying and more stable rela-
tionships in both the short and the long term.
Some of the positive effects Bortolotti mentions seem plausible, such as
having a strong sense of security and confidence in a relationship as a result
of partner idealization, or reinterpreting our partners’ weaknesses as
strengths. But I suggest that, when engaging in the ameliorative project,
Bortolotti’s approach is too one-sided, much like Velleman’s. While
Velleman focuses too much on the beloved and how they express them-
selves, Bortolotti focuses too much on the lover. She is focusing on the
optimistic agent and whether optimistic beliefs are good or bad for this
particular agent.
Since we are focusing on being better lovers, there is much more to
consider here. Loving people involves two (or more) people: at least a
lover and a beloved (this is even true of unrequited love, or loving very
young children). Bortolotti’s one-sided approach is probably due to focus-
ing on the effects of optimistic beliefs on psychological health: whether
true or false beliefs lead to psychological well-being or distress, that is,
psychological well-being or distress for the agent having the beliefs.
Bortolotti argues: “the belief that the partners share features with us and
with our ideal partners sustains our motivation to solve the problems our
relationship may be facing” (2018: 530). That might be true, but if we
want to discuss better love, we need a more nuanced approach. Or rather,
we need to add another standpoint. Bortolotti’s approach is much like
investigating whether Dante’s illusions of Beatrice are healthy for Dante’s
psychological well-being, or have positive effects for his take on their rela-
tionship (whatever relationship that might be). But those questions should
make us at least a bit uncomfortable, knowing there was another person
involved. It feels as if we’re asking the wrong questions—or rather, asking
not enough questions. What about Beatrice? What about her experiences,
beliefs, and psychological well-being?
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 77

Bortolotti continues: “In order for us to be successful agents in the face

of constant challenges, we need to believe that we can change things for
the better, and in order to do that we need to have a sense of competence,
control, and efficacy that propels us forward, a sense that our goals are
indeed desirable and attainable” (2018: 531). While this is not something
I want to argue against, I think that to ‘change things for the better’ we
should at least look at everyone involved, unless we only want to change
things for the better for no one but ourselves. Such a project would surely
not be what most people have in mind thinking about being better lovers.
When one tries to answer those missing questions, one would see that,
while the positive illusions might be beneficial for Dante, they are proba-
bly not that beneficial for Beatrice. Dante’s positive illusions paint such an
unrealistic picture of Beatrice that it silences her real self. Beatrice has no
voice: we don’t know who she is, what she experiences, or what she wants.
Focusing too much on the beneficial consequences of positive illusions
for the lover would neglect the process of attending to the beloved.
Bortolotti argues, by focusing on the loving agent, that positive illusions
could make for stable relationships by giving the agent the motivation to
work for that relationship. Bortolotti’s account thus seems to be more
about stable relationships than about being better lovers. And while those
could influence each other, it is certainly possible to have a stable relation-
ship without being good lovers, for example, a very oppressive, but stable,
relationship.
Besides a too narrow focus, Bortolotti mentions one particular effect of
optimistic beliefs that could turn out as not just meaningless love, but
harmful for the beloved. The effect furthermore shows the need for love
as actively attending to real people. Sandra Murray and colleagues discuss
an effect of partner idealization that is labeled as reflective appraisals: when
partners are idealized, they come to see themselves as we see them, and
live up to our high standards (Murray et al. 1996b). Based on this view,
Bortolotti argues that positive illusions can bring success in romantic rela-
tionships not because perceptions of the romantic partners are realistic,
but because they have a positive effect on our behavior in the relationship
and support us in the pursuit of our relationship-related goals when prob-
lems emerge (2018: 527). “Intimates can actually turn self-perceived frogs
into the princes or princesses they perceive them to be” (Murray et al.
1996b: 1158). But it seems to depend heavily on the nature of the illusion
whether such a process of adapting to others’ expectations is necessarily a
good thing. If men having the illusions of thin, hourglass-shaped, hairless
78 L. SPREEUWENBERG

women leads to women living up to those expectations, adapting to your

lover’s illusions could be quite harmful (see the passage about The Beauty
Myth above). Murray et al. found that, over time, the idealized evaluations
became more realistic, not because people experienced disappointment
and lowered their expectations accordingly, but because partners rose up
to the challenge and exhibited the qualities that were initially attributed to
them. As a result of reflective appraisals, the gap between idealization and
reality shrinks. But this is not necessarily a good thing, considering the
particular aforementioned example of self-masking or self-destructing
behavior, only to live up to expectations.
Furthermore, the mechanism of illusions turning into beauty standards
seems to be about non-moral ideals,5 but what about illusions about
someone’s moral character? Reflective appraisals could turn out great
when we expect someone to be good: what could be wrong with someone
trying to live up to expectations concerning moral character? Here, again,
it seems to depend heavily on the nature of the illusion whether the pro-
cess of adapting to others’ expectations could be harmful. Depending on
the society we live in, there are particular expectations of, for example, a
‘good woman’. While the dominant expectations fortunately no longer
include role patterns of wives cooking and husbands breadwinning, many
other unequal patterns and expectations are still at play. Kate Manne
points out roles and standards of Western society, in which women are
considered moral givers (2017). A ‘good girl’ gives, doesn’t ask for any-
thing, is expected to be grateful, owes things to others as opposed to
being entitled to something—especially ‘moral goods’, such as attention,
care, sympathy, respect, and admiration. Manne further explains that we
must understand misogyny as a characteristic of such social environments,
in which women are susceptible to hostility due to the maintenance of
these expectations: women are considered ungrateful, sour, aggressive, or
worse when they don’t live up to these expectations. If these kinds of ide-
alized evaluations become more realistic because partners rise up to the
challenge and exhibit the qualities that are attributed to them, reflective
appraisal is (in this case) a harmful dynamic—a power play by which
suppressed groups continue to be suppressed by ongoing fantasies of the
oppressor.

5
Although Heather Widdows makes a compelling case for beauty being considered an
ethical ideal in her book Perfect me (2018).
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 79

This dynamic could be harmful for men, too: it is not too hard to come
up with examples of harmful fantasies that concern men: ‘men never cry’,
‘men are sexual predators’. Moreover, it is not solely a gender issue, but
something that concerns all forms of inequality, stereotypes, and social
patterns: ‘black women are hypersexual’, ‘black men are dangerous’,
and so on.
A meaningful connection in which ‘two people have the right to use the
word love’, as Rich put it, should look past stereotypes, social patterns,
and self-serving fantasies. Only by attending to the particular and unique
around us, we can let go of our self-serving worlds. Murdoch herself did
not advocate political use of loving attention. However, loving attention
would make us better lovers in general and enable us to live better with the
real people around us. Fantasies are only given room to grow when we
keep someone (or a certain group) out of our sight or at an emotional
distance. Velleman would probably not argue that loving fantasies is par-
ticularly valuable, but his account of love is not equipped to prevent or
combat idealized fantasies. Loving, as both described by Murdoch and
argued for here, focuses on eliminating these fantasies—adapting our con-
cepts to the uniqueness of the different people we meet, by looking at the
world around us. Boxes, categorizations, and stereotypes make way for
real people.

4.6   The Ongoing Practice: Getting to Know

Versus Knowledge
I have discussed several distinctions between Velleman’s and Murdoch’s
accounts of love and argued that a Murdochian kind of loving is better
suited for the ameliorative project. Love should be more than contingent
and requires actively looking. Furthermore, loving should go beyond see-
ing another’s dignity: it is precisely looking at the particular and unique
that helps us move away from our ego, a practice we need to escape self-­
serving fantasies and to see the reality of others.
There is another important aspect to this. To love better is to engage in
an ongoing practice. We are successful when we engage in this activity, not
when we reach a particular goal such as a contingent fit or valuing a per-
son’s dignity. Love is a process, a progressive attempt. Really looking at
others is not easy. It is ‘the hard way’, writes Rich. Are we ever able to see
someone’s truth? Are we not always bound by the concepts we have and
80 L. SPREEUWENBERG

therefore never able to really see someone? And if so, what is ‘better love’
exactly aiming at? What if Dante actively engages in an activity of focused
loving attention but still fails to see some aspects of Beatrice’s truth? And
how well do we have to know each other to speak about love?6
Seeing our loved one’s reality must not be mistaken for knowing every-
thing about them or even understanding them. ‘Better love’ as it is argued
for here, does not entail that we know or need to know every detail about
our beloved. Carla Bagnoli (2018) rightly points out that sometimes
understanding might be too violent a modality of relating to others, “like
poking into their private reality, rather than simply accepting their alterity
and respecting their opacity” (p. 82). Love does not aim at completely
understanding others. Love has no ‘end goal’, which would entail particu-
lar knowledge about the beloved. Loving, on this account, indeed does
not aim, but is the process of getting to know others. It is an ongoing
practice, being perceptive of others as they are.
Murdoch knew perfectly well that we are never really able to see reality
successfully. We can only look through our own eyes, with our own con-
cepts, culture, and history. Perception is therefore a restricted capacity:
what we see depends on what particular concepts we have, and if we don’t
have (or acquire) the right conceptual resources, we might be forever
blind to some particular parts of the truth. It might be impossible for
Dante to know Beatrice’s reality. For example, some of her experiences
will always remain hidden to Dante to a certain extent: he will never know
what it feels like to give birth. If we cannot succeed in seeing truth, why
would being open to this truth be a better way of loving?
With the example of M and D Murdoch wanted to imagine a case in
which the reader could feel approval of M’s change of view. She also admits
(the example is especially designed that way) that in real life it might be
very hard to decide whether M’s ultimate judgment of D in the example
is morally appropriate or not (Murdoch 1971: 17). The reader doesn’t
know D, and therefore cannot evaluate whether D is really a good-hearted
girl, or whether M’s loving attention leads M either to see truth or just to
see more fantasies. There is not enough space here to elaborate extensively
on Murdoch’s view of morality or her meta-ethical perspective. But what
happens in the example of M is important. It is used to show that moral
activity can happen in the inner life, and Murdoch positions herself against

6
This is a question Eileen John asks in her paper “Love and the Need for
Comprehension” (2013).
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 81

the “existentialist-behaviorist types of moral psychology” (p. 9) who claim

that “mental concepts must be analyzed, genetically and so the inner must
be thought of as parasitic upon the outer” (p. 10). On such a view M’s
change of mind about D is an empty activity, because no kind of outer
structure is present. But Murdoch wants to show that M is engaged in an
internal struggle and that her activity feels very familiar. Furthermore, M’s
activity is peculiarly her own: she could not do this thing in conversation
with another person, so the quasi-scientific notion “anything which is to
count as definite reality must be open to several observers” cannot be
applied to this example (p. 22). Murdoch’s point is that what M is doing,
that is, focusing her attention outside herself, counts as a moral activity.
Murdoch even puts this activity, that is, love, at the center of morality.
Murdoch’s entire concept of love and morality is a concept of progres-
sion: loving attention (and being good) is “infinitely perfectible” (p. 23).
Loving, or the picture she paints of it, has built in the notion of a necessary
fallibility. M is attached to the concepts she has and can do nothing but try
to see the truth.

M is engaged in an endless task. As soon as we begin to use words such as

‘love’ and ‘justice’ in characterizing M, we introduce into our whole con-
ceptual picture of her situation the idea of progress, that is the idea of per-
fection: and it is just the presence of this idea which demands an analysis of
mental concepts which is different from the genetic one. (Murdoch 1971: 23)

What is at stake, says Murdoch, is the liberation of morality and phi-

losophy as a study of human nature, from the domination of science
(1971: 26). Morality is less to do with the isolated will jumping in and out
of an impersonal logical complex, and more with the progressive attempt
to see a particular object clearly.
In The Sublime and the Good, Murdoch, by juxtaposing Kant and Hegel,
argues that seeing that object—‘the sublime’—clearly is always a progres-
sive attempt (Murdoch 1997). Kant thinks of the sublime as not given but
only vaguely adumbrated by reason, only occasioned by natural objects
(non-historical, non-social, non-human), a systematic perception of nature
in which time, place, and the nature of our sensibility play no part (1997:
49). Hegel, on the other hand, makes social and historical and human and
concrete what Kant has offered as abstract, non-historical, and so forth:
the unity of the ethical substance is given. For Murdoch, both are wrong
(or both are somewhat right!). For Murdoch, there is, contra Hegel, an
82 L. SPREEUWENBERG

abstract essence of morality, which she names the Good. Contra Kant, we
can only perceive this essence within our own place, time, and with our
own eyes, she argues.

Do we really have to choose between an image of total freedom and an

image of total determinism? Can we not give a more balanced and illuminat-
ing account of the matter? I suggest we can if we simply introduce into the
picture the idea of attention, or looking, of which I was speaking above. I
can only choose within the world I can see, in the moral sense of ‘see’ which
implies that clear vision is a result of moral imagination and moral effort.
(Murdoch 1971: 35–36)

Murdoch argues that moral tasks are characteristically endless, not only
because, within a concept, our efforts are imperfect, but also because as we
move, really look and open up, our concepts themselves are changing
(1971: 27). “M’s independence of science […] rests not simply in her
moving will but in her seeing knowing mind” (p. 27). Love is getting to
know the individual, and M confronted with D has an endless task. At the
end of the example, M sees D as ‘spontaneous’ and ‘delightfully youthful’.
But since M has an endless task, the example might as well go on while M
continues to look upon D with loving attention. It might be the case that
M later alters her view of D from ‘spontaneous’ to ‘somewhat impulsive’,
for example, after continuously looking at D without letting her own
desires and needs play a part. This imaginative continuation of the exam-
ple also shows that loving attention is not meant to be ‘judging everything
as positive’; it is about trying to see the Good as an obedience to reality,
while knowing that we could never fully grasp that reality. Even though
Dante will never fully grasp Beatrice’s reality, opening his eyes, engaging
in loving attention, focusing on Beatrice as particular and unique beyond
his ego, is engaging in moral activity. It is not the facts, the outer activity,
or mental concepts that can be analyzed that matter morally. It is the inner
activity, the effort of directing our attention on individuals, of obedience
to reality as an exercise of love. Murdoch suggests that ‘reality’ and ‘indi-
vidual’ present themselves to us in moral contexts as ideal end-point, an
end-point imperfect humans cannot ever reach, but must aim at. “This
surely is the place where the concept of good lives. ‘Good’: ‘Real’: ‘Love’.
These words are closely connected”, writes Murdoch (1971: 41).
The value of using the concept of love as an ongoing practice is thus
that we have to adjust our concepts constantly to the reality outside us.
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 83

Dante is never done getting to know Beatrice; never will he have an

amount of knowledge of Beatrice that is a particular end-goal. The value
of the practice of love is not in the knowledge we have of our beloveds,
but in the ongoing effort to get to know each other. Dante could argue
that he knows Beatrice, but he is not able to move away from his ego, not
able to get rid of his harmful self-serving fantasies, unless he engages in an
ongoing effort of really looking at her.
Suppose Dante would engage in loving practice and learn that Beatrice
is interested in mathematics, is a big fan of hip-hop, and has a birth mark
on her left forearm. Although he now has some knowledge of Beatrice,
the moment Dante quits this practice, the dangers of the ego are lurking.
Firstly, he doesn’t learn more about Beatrice, and harmful fantasies about
the things he doesn’t know about her yet—for example, ‘women effort-
lessly having no body hair’—could prevail. Secondly, Beatrice could
change—for example, changing her musical preference from hip-hop to
neo-soul—and Dante would be stuck with an old image of her, back again
at being blind for the real Beatrice. The ongoing aspect of love as a prac-
tice is important to ensure that we don’t fall back on fantasies. Velleman’s
account is not equipped to improve the way we love, in this sense too,
because he fails to incorporate this ongoing activity.
The phrase ‘love as a practice’ already reveals something about the
nature of this love considering knowledge of the individual. Knowledge
seems to be the result, while ‘getting to know something or someone’ is
the activity, the progressive attempt Murdoch describes as love. To be bet-
ter lovers is not about what or how much knowledge we have of each
other, but rather about the practice of being open to others. Loving is
attending, not the knowledge we get out of this attention. Surely we do
want our lovers to know something about us, but there’s no threshold of
how much we should know about our beloveds to speak of love in this
sense. In her paper “Love and the Need for Comprehension”, Eileen John
defends the possibility of love with failure of comprehension (2013). She
argues that the beloved person has a say as to the kind and depth of knowl-
edge that is required to count as loving him or her. The beloved has to be
able to recognize herself in the love (2013: 286), and this involves
acknowledging the beloved as a being with consciousness, interests, and
authority over what is important to her. This might solve Beatrice’s prob-
lem: Dante’s love would not count as love because Beatrice probably
would not feel acknowledged as herself. But it does not fix the problem of
reflective appraisals: fantasies lead to (harmful) expectations that many
84 L. SPREEUWENBERG

beloveds want to live up to. In the light of this mechanism, it seems strange
to ‘put the burden’ of whether someone’s love counts as love on the
beloved person (having a say as to the kind and depth of knowledge is
required to count as loving her), while at the same time, the beloved
adapts—or feels pressure to adapt—the qualities or interests of what the
knowledge is about to the fantasies of the lover. I suggest that we should
not look at thresholds for knowledge to speak of love, but rather at pro-
gressive attempts. It is not the knowledge that others have of us that makes
love meaningful. It is the effort of a lover willing to put their ego aside and
open up, the ongoing activity of really looking at us.
Furthermore, the knowledge our lovers have of us is always colored by
the particular concepts they have. A lover might know that I’m a philoso-
pher, but her concept of what that exactly entails depends on her concepts
and experiences—her frame of reference. What is so particular meaningful
in loving is that it is a constant attempt to look past our own egos and see
the people and things around us. My lover constantly adapts her concept
of what a philosopher is by continuously looking at me and my experi-
ences. Loving forces us to adjust what we know to what we see, but since
what we see depends on what we already know, the activity is progressive
and infinitely perfectible. If we keep attending to everything outside us,
we can come to know more and more about the world and the people
around us.
This is not to say that we should accept everything we see or work
toward that: really looking could reveal things we should not be accepting.
Loving attention precisely is able to discern immorality and inequality by
‘seeing someone’s truth’, as well as discovering patterns and see social
structures as these are part of the particular person we are attending to
(Clarke 2012). Attending to someone could on the one hand reveal some
possibly good character traits or explanations of character, but it will also
reveal that person’s immoral actions and character traits we disapprove of.
Loving in this sense is the opposite of blindly embracing: really looking
can be eye-opening.
We only have our own eyes to look with, our own backgrounds, cul-
ture, upbringing, and so forth. It is inevitably difficult to form connec-
tions when we notice differences between ourselves and others. It might
take much longer before we are able to really look at those who are differ-
ent and it might take a while before we can even catch a glimpse of their
reality. Some groups or experiences are almost invisible because of, for
example, non-appearance in the media or science. Love as attending to the
 4 ‘LOVE’ AS A PRACTICE: LOOKING AT REAL PEOPLE 85

things outside ourselves could therefore not only make us better (roman-
tic) lovers, but make us form meaningful connections with people of dif-
ferent genders, skin colors, sexual preferences than our own. Are we
willing to escape our self-serving egos? If we are willing to put in some
effort and engage in an ongoing activity of opening up to each other, to
put our ego aside trying to see each other’s truths, we could all be pro-
gressively better lovers.

References
Abramson, Kate, and Adam Leite. 2011. Love as a Reactive Emotion. The
Philosophical Quarterly 61 (245): 673–699.
Alighieri, Dante. 1294. Vita nuova.
Bagnoli, Carla. 2003. Respect and Loving Attention. Canadian Journal of
Philosophy 33 (4): 483–515.
———. 2018. Constrained by Reason, Transformed by Love: Murdoch on the
Standard of Proof. In Murdoch on Truth and Love, 63–88. Cham: Palgrave
Macmillan.
Baker, Lynn A., and Robert E. Emery. 1993. When Every Relationship is Above
Average. Law and Human Behavior 17 (4): 439–450.
Bortolotti, Lisa. 2018. Optimism, Agency, and Success. Ethical Theory and Moral
Practice 21 (3): 521–535.
Buunk, Bram P., and Regina J.J.M. van der Eijnden. 1997. Perceived Prevalence,
Perceived Superiority, and Relationship Satisfaction: Most Relationships are
Good, But Ours is the Best. Personality and Social Psychology Bulletin 23
(3): 219–228.
Clarke, Bridget. 2012. Iris Murdoch and the Prospects for Critical Moral
Perception. In Iris Murdoch: Philosopher, ed. Justin Broackes, 227–253. Oxford:
Oxford University Press.
Fowers, Blaine J., Eileen Lyons, Kelly H. Montel, and Netta Shaked. 2001.
Positive Illusions About Marriage Among Married and Single Individuals.
Journal of Family Psychology 15 (1): 95.
Frankfurt, Harry. 2009. The Reasons of Love. Princeton University Press.
Friedan, Betty. 1963. The Feminine Mystique. New York.
Grenberg, Jeanine M. 2014. All You Need is Love? In Kant on Emotion and Value,
210–223. London: Palgrave Macmillan.
Harcourt, Edward. 2009. Velleman on Love and Ideals of Rational Humanity. The
Philosophical Quarterly 59 (235): 349–356.
Haslanger, Sally. 2012. Resisting Reality: Social Construction and Social Critique.
Oxford University Press.
hooks, bell. 2000. Feminist Theory: From Margin to Center. Pluto Press.
86 L. SPREEUWENBERG

———. 2001. All About Love: New Visions. Harper Perennial.

John, Eileen. 2013. Love and the Need for Comprehension. Philosophical
Explorations 16 (3): 285–297.
Jollimore, Troy. 2011. Love’s Vision. Princeton University Press.
Keller, Simon. 2000. How Do I Love Thee? Let Me Count the Properties.
American Philosophical Quarterly 37 (2): 163–173.
Kolodny, Niko. 2003. Love as Valuing a Relationship. The Philosophical Review
112 (2): 135–189.
Lopez-Cantero, Pilar. Manuscript. Loving a Narrator.
Manne, Kate. 2017. Down Girl: The Logic of Misogyny. Oxford University Press.
Merritt, Melissa McBay. 2017. Love, Respect, and Individuals: Murdoch as a
Guide to Kantian Ethics. European Journal of Philosophy 25 (4): 1844–1863.
Millgram, Elijah. 2004. Kantian Crystallization. Ethics 114 (3): 511–513.
Milligan, Tony. 2013. Valuing Love and Valuing the Self in Iris Murdoch.
Convivium: Revista de Filosofia.
Murdoch, Iris. 1971. The Sovereignty of Good. Routledge.
———. 1997. The Sublime and the Good. In Existentialists and Mystics: Writings
on Philosophy and Literature, ed. Peter Conradi. New York: Penguin.
Murray, Sandra L., John G. Holmes, and Dale W. Griffin. 1996a. The Benefits of
Positive Illusions: Idealization and the Construction of Satisfaction in Close
Relationships. Journal of Personality and Social Psychology 70 (1): 79.
———. 1996b. The Self-Fulfilling Nature of Positive Illusions in Romantic
Relationships: Love is Not Blind, But Prescient. Journal of Personality and
Social Psychology 71 (6): 1155.
Rich, Adrienne. 1995. On Lies, Secrets, and Silence: Selected Prose 1966–1978. WW
Norton & Company.
Rusbult, Caryl E., Paul A.M. Van Lange, Tim Wildschut, Nancy A. Yovetich, and
Julie Verette. 2000. Perceived Superiority in Close Relationships: Why it Exists
and Persists. Journal of Personality and Social Psychology 79 (4): 521.
Smuts, Aaron. 2013. In Defense of the No-Reasons View of Love.
Velleman, J. David. 1999. Love as a Moral Emotion. Ethics 109 (2): 338–374.
Widdows, Heather. 2018. Perfect Me. Princeton University Press.
Wolf, Naomi. 1991. The Beauty Myth: How Images of Beauty are Used Against
Women. Random House.
Zangwill, Nick. 2013. Love: Gloriously Amoral and Arational. Philosophical
Explorations 16 (3): 298–314.
 CHAPTER 5

Love, Choice, and Taking Responsibility

Christopher Cowley

There is a long-standing philosophical discussion about the relationship

between love and choice. The most simplistic versions (in the case of
romantic love and friendship between adults) see love as something one
“falls” into, without any choice at all or (in the case of family love) as
something that one grows up with and grows into, again without choice.
I certainly do not want to deny this important passive element. A more
sophisticated account of love would accommodate some degree of indi-
rect choice. For example, I can feel an initial interest in a person and choose
to seek her out more, thereby improving the chances that love will develop.
I can also choose to create and sustain the conditions that support love,
for example, by avoiding infidelity and long commutes. However, indirect
choices, as exemplified above, are relatively discrete or time-limited. In
contrast, I want to explore a more temporally extended kind of choice,
which starts when I take responsibility. I will argue that one essential com-
ponent of longer-term adult romantic and friendly love requires the lover
to take prospective responsibility—as a kind of attitude—for meeting the
beloved’s (friend’s) unpredictable and possibly onerous needs in the

C. Cowley (*)
School of Philosophy, University College Dublin, Dublin, Ireland
e-mail: christopher.cowley@ucd.ie

© The Author(s), under exclusive license to Springer Nature 87

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_5
88 C. COWLEY

future. This kind of attitude is sometimes described as sensitivity or a kind

of attention, but I do not believe these concepts are robust enough to
capture the phenomenon.

5.1   The Little Prince

My starting point is a book ostensibly for children, but also a perennial
favorite among adults: St Exupery’s The Little Prince (2001/1943). The
Prince is visiting the earth from his home planet and meets a fox. The
Prince is lonely and wants to play. “I can’t play with you. I’m not tamed,”
says the fox. “What does ‘tame’ mean?” asks the Prince. Taming means to
“establish ties,” says the fox, and continues:

To me, you are still nothing more than a little boy who is just like 100,000
other little boys. And I have no need of you. And you, on your part, have no
need of me. To you, I am nothing more than a fox like 100,000 other foxes.
But if you tame me, then we shall need each other. To me, you will be
unique in all the world. To you, I shall be unique in all the world.

Later on, the Prince talks to a rose and compares the earthly rose to his
own rose back on his home planet.

You are beautiful, but you are empty. […] To be sure, an ordinary passer-by
would think that my rose looked just like you − the rose that belongs to me.
But in herself alone she is more important than all the hundreds of you
other roses: because it is she that I have watered; because it is she that I have
put under the glass globe; because it is she that I have sheltered behind the
screen; […] Because she is my rose.

He eventually returns to the fox to say goodbye, and the fox tells him:
“you become responsible, for ever, for what you have tamed. You are
responsible for your rose.”
I have to address a translation issue right away. The verb and adjective
“tame” has unfortunate connotations in English, which would seem to
make the word inappropriate in any discussion of mature, mutually
respectful adult love. After all, the Taming of the Shrew has always been
problematic in this respect. But the etymology of the French word app-
rivoiser is revealing: it comes from the Latin “to make private.” This is
much richer than merely taming or domesticating the wild animal, let
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 89

alone the wild woman. Making private means bringing another person
into the privacy of one’s own life. Understood this way, taming can be
symmetrical and non-oppressive. It could refer not only to the adaptations
and compromises necessary to share a household (and a bed) with another
person, but also to the mutual attunement, to the partial sharing of ends
and projects, and to the attenuation of interpersonal barriers to the point
of developing the first-person plural pronoun; at that point you and the
other person have a joint privacy and can confront the world as a “we.”
And as the fox points out, once you lovingly bring a person into your pri-
vate lifeworld, you have taken responsibility for them. This central role for
responsibility is what I want to explore.

5.2   Responsibility
Unfortunately I am defining one polyvalent term, “love,” in terms of an
equally polyvalent term, “responsibility”! So I need to make distinctions.
Perhaps the most obvious form of responsibility is retrospective and has to
do with answerability for past actions. “Who is responsible for this mess?”
probably means two things: (1) who is causally responsible, that is, “who
did it (or failed to prevent it)?” and (2) who is morally responsible, in the
sense of “who is an appropriate target for blame?” This kind of responsi-
bility is most forcefully on display in the criminal justice system. But it is
not my main concern here.
A second meaning of responsibility is prospective. I am responsible for
something in the future. There are two versions of this, which I will call
“closed” and “open.” The closed form of prospective responsibility con-
cerns a specific duty attached to a role. A job description typically com-
prises a list of responsibilities. “Who’s responsible for after-sales support?”
“Oh, that’s Smith.” Customers will have certain typical needs or com-
plaints after sale, and Smith is ready—into the future, so long as she con-
tinues to fill the role—to carry out certain procedures (e.g. warranty-backed
refunds) to meet these needs and resolve these complaints, in accordance
with her employment contract, with company policies, and with business
and legal norms. The cost and the risks of after-sales support are built into
the business plan and the original purchase price.
In contrast, open prospective responsibility involves a readiness to deal
with someone else’s future needs and wants that are much less predictable
than those of the customer. And they may be unpredictable not only in
terms of the precise content, but also in terms of their onerousness on me
90 C. COWLEY

(and their incompatibility with my other future, partly unknown commit-

ments). When I am prospectively responsible in this more open sense, I
might not be able to confidently imagine much about my future interac-
tions with the other. And yet despite the risks and uncertainty, this is the
sort of responsibility that the fox is talking about. Insofar as the fox is my
friend and I love it, then I make myself available to the fox into the
future—I take responsibility for it.1
The word “responsibility” obviously comes from the concept of
“response.” In the retrospective sense, holding someone morally respon-
sible means I am asking them to respond appropriately to the damage they
have caused, for example, with apology or compensation. In the prospec-
tive sense, taking responsibility for after-sales support means a commit-
ment to respond to the customer’s predictable future needs insofar as such
needs relate to their legitimate purchase of our company’s products or
services.
However, responding appropriately to the unpredictable future needs
of a beloved or a friend is much riskier. It’s all very well for me to tell my
friend sincerely: “whatever you need, just give me a call”; but when that
friend starts to need me too much, then it will strain the friendship. With
the after-sales service, there might come a point where the demanding
customer can be dismissed or ignored: “we have helped them as much as
could reasonably be expected, and we have no further legal or moral obli-
gation.” Alternatively, if I find the job of after-sales support simply too
onerous, I can always quit my job without reproaching either myself or my
employer. I can say to myself: “I did not understand the responsibilities I
was signing up for, now I do, and it’s just not for me.”
But such a “graceful exit” is much more difficult when a friendship is
tested by unexpected asymmetrical needs. Some friendships are deepened
by such testing; others fall apart, and often it will be a matter of luck
whether a particular friendship is tested in a constructive or destructive
way. (1) If my friend’s future needs had been a little more predictable or a
little more typical, then I might have been able to make a more informed
decision about the degree of friendship I wanted to offer at the start; (2)
if the needs had been a little lesser, or if my competing commitments had

1
Although note that in the original French, the fox is le renard, in other words masculine.
English relegates most animals to “it,” and this already hinders the possibility of friendship.
To take responsibility for an “it” is quite different from taking responsibility for a “him.”
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 91

been a little lesser, or if others had been able to help, then I would have
been able to help her.
Some breakdowns can be a matter of sadness without bitterness, espe-
cially when there is evidence of a non-culpable change of personality,
interest, or circumstances. However, my inability or unwillingness to
respond to the other’s unexpected needs can lead to a bitterness in her
that hardens into a sense of betrayal, perhaps to the point of retrospective
redefinition: “I guess you were never really a friend at all,” she tells me.
And I have to acknowledge that she might be entitled to think this, just as
I have to acknowledge the lameness of my self-consoling thought that my
past commitment had been sincere. I will return to the issue of retrospec-
tive redefinition erelong.
Throughout, I may see myself as lacking choice, in accordance with the
simplistic passive model of love. This person and I happened to feel affec-
tion for one another when we met, we became friends, and we took
responsibility for one another. But later, when she presented her unpre-
dictable and burdensome needs on me, I discovered (passively) that my
affection was not up to the task. I blame bad luck. However, if it really
were only a question of mere bad luck and passivity, then I would not be
left with such a bad feeling about it. There has to be more room for
my agency.

5.3   Taking Responsibility

I have already hinted at two kinds of open responsibility-taking, which I
can now distinguish explicitly as “easy” and “hard.” When I first meet the
person, and we like each other and become friends, then part of what it
means to become a friend is to take open prospective responsibility for her
unpredictable future needs. Acting on affection, it is relatively easy to say
“whatever you need, give me a call.” Later, when we have been friends for
a while, my friend develops a need which is surprising and onerous. At this
stage I can take “hard” responsibility by making larger adjustments to my
life in order to meet her need and steeling myself for future needs (which
may now be more predictable, but may not). In so doing, our friendship
has been tested, and, if I am able, I can choose to deepen it. Actually, it
would be better to speak of the comparative—“harder” rather than
“hard”—since I may be tested even more in the future. At any rate, it is
the hard responsibility-taking that interests me, the one that makes or
breaks the friendship.
92 C. COWLEY

(Again I emphasize the onerousness. If the friend’s need is surprising

but not onerous, then I can meet it easily under the scope of the original
open prospective relationship-taking; our friendship has not really been
tested, and I have faced no difficult choice.)2
In most discussions of responsibility, taking is reducible to being. If I
am retrospectively responsible for spilling the milk, then I ought to take
responsibility for it now by cleaning it up. Alternatively, I might refuse to
take responsibility because I do not see myself as being responsible, and
we may argue about the basis for my alleged responsibility—but I will still
agree that if I were responsible, then I ought to take responsibility.
There is also a more interesting form of taking retrospective responsi-
bility, described by Susan Wolf as a “nameless virtue,” something akin to
generosity. Her discussion is entirely about retrospective responsibility,
but the basic idea will be useful for my discussion of prospective responsi-
bility as well.

It involves living with an expectation and a willingness to be held responsi-

ble for what one does, understanding the scope of “what one does,” par-
ticularly when costs are involved, in an expansive rather than a narrow way.
It is the virtue that would lead one to pay for the vase that one broke even
if one’s fault in the incident was uncertain; the virtue that would lead one to
apologize, rather than get defensive, if one had unwittingly offended some-
one or hurt her. (Wolf 2004: 121)

Wolf’s point is that we have to acknowledge the “messiness and the

irrational contingencies of the world” (p. 122). In seeking to calculate
blameworthiness too precisely, we might withdraw from “the only ground
on which it is possible for beings like ourselves to meet” (ibid.). In short,
when there are certain kinds of doubt about whether I am retrospectively
responsible, it is virtuous to take responsibility.3
Wolf recognizes the obvious risk of exploitation: a woman in an abusive
marriage who is too inclined to “apologize, rather than get defensive”; the

2
My account also allows that my partner might be mistaken about her needs. Even if I am
able to meet her needs, I might not do so if I judge that it is not in her interest. This should
not be a matter of my judgment and my decision, but part of an ongoing conversation
between two concerned equals.
3
It is interesting that Wolf argues in terms of a virtue rather than a choice; it suggests that
the person already possesses the virtue, and therefore does not have to make a difficult
choice, since her virtuous disposition will make it easy.
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 93

office “mug” who takes responsibility for the broken vase because of a
misguided gratitude for being hired. For the moment, if we limit the dis-
cussion to non-exploitative relationships between social equals, then Wolf
has pinpointed an important virtue not only relevant between strangers
but especially between loving friends. To put it another way, an ongoing
friendship requires trust; when I realize that I have offended the friend,
then my friendship appropriately inclines me to trust that she has good
reason to be offended, even if I cannot (yet) see what I have done wrong—
and so I take responsibility and apologize.4
What would it mean to take prospective responsibility in Wolf’s quasi-­
generous sense? At first glance, this might mean no more than volunteer-
ing for after-sales customer care, as one of a range of extra tasks open for
ambitious employees. More complexly, one can imagine something like
vicarious liability in employment law: when my employee breaks some-
thing, then I as her employer become straightforwardly liable for the dam-
ages, even if I took all reasonable care in training and supervising the
employee, and even if there was no reasonable way that I could have antic-
ipated or prevented the breakage (i.e. I was not legally negligent).
But friendship is philosophically fascinating because of the absence of
formal structures. There is a strong sense that—within broad limits of
intelligibility about whether it is a friendship at all—it is up to us what hap-
pens to our friendship, whatever third parties might admire or criticize in
us. And what you and I have made of our friendship up to now will limit
the options available to me for the future, including the decision to take
or not to take responsibility for your unexpected and onerous needs when
they present themselves.
I spoke earlier about retrospective redefinition, where one friend’s
abandonment of another is taken as revealing not only the present state of
4
This would relate to Niko Kolodny’s (2003) influential conception of love as being based
on “relational reasons.” Kolodny was rejecting the prominent “reasons-conception” of love.
According to this conception, in order to be intelligible, love has to be based on reasons
generated by the beloved’s properties: X loves Y because of Y’s properties A, B, C. Two
weaknesses of this conception are that (1) it cannot accommodate love that continues despite
Y’s loss of some of the relevant properties; and (2) it cannot accommodate Y’s individuality,
since in principle Y could be replaced by Z who has the same properties, or even improved
versions of such properties, and X would be rationally committed to diverting his love to
Z. Kolodny’s account deals with both problems by arguing that there are such things as
reasons for love, but that they lie in one’s prior relationship with X, that is that she is already
my romantic partner or my friend, and that fact gives me some non-overriding reasons to
continue loving her.
94 C. COWLEY

the friendship, but of the whole friendship, all along. This process can also
work in the opposite direction: one friend taking responsibility for anoth-
er’s unexpected onerous needs does not so much deepen the preexisting
friendship, but reveals the depth that the friendship had all along. The
actual contained the hitherto unknown potential to collapse or to deepen.
Even when there are stages in the history of a friendship, it has an irreduc-
ible narrative wholeness that remains vulnerable to future developments.
Even if this invites a deterministic reading, some of the past meaning still
lies within my present choice.
One might even be inclined to take responsibility for a presently
estranged ex-friend, precisely in order to preserve the original friend-
ship—in a spirit of “for old times’ sake”—without any desire to resume
the friendship. It is the original friendship that creates a life-long obliga-
tion, regardless of whether we have drifted apart culpably or non-culpably
in the meantime, and regardless of whether I do not like or respect what
my ex-friend has turned into. Again, in the words of the fox: “you become
responsible, for ever, for what you have tamed.”
In arguing for this conception of choice within friendship, I am draw-
ing on a debate between different conceptions of well-being. One concep-
tion is sequential: I live through a series of moments, in each of which my
well-being is based on some time-indexed objective state of affairs. If I am
happy at time t1, nothing can change the value of the discrete fact of my
happiness at that moment, even if my subsequent access to it at t2 (when
I am unhappy) is only through more-or-less reliable memory.5 The con-
trasting view, instantiated by the Ancient Greek concept of eudaemonia, is
holistic rather than sequential. Only an entire life can have an objective
determinate value, and that value can only be fully apprehended at the
end. Even if I seem to be happy during an episode, the full value of that
episodic happiness will depend on the place of that episode in the story of
my life, including among unknown future events. For example, when I
successfully land a permanent academic job, my life seems to acquire
objective well-being. When I learn that a much more deserving and needy
applicant was rejected for the same job, my overall well-being is
undermined.

5
In Gershwin and Gershwin’s song “They can’t take that away from me,” originally writ-
ten for the 1937 Fred Astaire movie Shall We Dance, the narrator speaks of his beloved’s
attractive and distinctive properties (e.g. “the way you wear your hat”). Even if cruel fate can
take her away from me, they cannot take away my memories of her, nor can they devalue the
quality of my remembered happiness with her.
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 95

5.4   The Marital Vow

Earlier I drew the distinction between “easy” and “hard” prospective
responsibility-taking. At the start of a relationship it is easy (or easier) to
take responsibility for the other’s unpredictable onerous needs. Because
friendship is so amorphous, it will help to consider an institution with a
certain traditional definition, launched by a formal expression of early pro-
spective responsibility-taking: the sincere marital vow. (Sincerity is not
enough, of course, if the protagonists are too young and ignorant. So I
shall be assuming two people who have known each other for a while, and
who are old enough—maybe 25?—to know something of the world, and
something about themselves, such as what they want and what they’ll set-
tle for.) I stress that it is the formality of the marital vow that interests me;
I do not consider marital love to be so distinct from friendly love when it
comes to the central phenomenon of taking responsibility. Most of what I
say in the following text could apply to friendship.
One form of the unconditional commitment might be to “love, honor
and cherish” the other “as long as you both shall live.” But the precise
words are less important than the unconditionality. Although statistically,
in some parts of the world, there might be as much as a 50–50 chance that
this couple will collapse or divorce, this couple sincerely and seriously
believes that (1) their union will not (those statistics concern other peo-
ple), and (2) they are determined to make it work, whatever it takes. What
makes this possible is a clear sense of shared identity that launches the
relationship. When something bad happens to her, or when she does
something bad, there is a clear sense that it happens to me, or that I have
done it. Even when we do not entirely share the same ends, projects, and
interests, the fact that they are her ends, projects, and interests makes them
mine as well.6
This notion of “whatever it takes” is interesting, because the two part-
ners almost certainly do not know, at the time of making the uncondi-
tional vow, what it will in fact take to keep them together. When I make
the unconditional vow to this person, I take responsibility not only for this
person and her unpredictable needs next week or next year, but also for
the person she will become, together with the unpredictable needs that she

6
David Enoch (2012) argues that identity-conferring commitments ground a sense of
“penumbral” agency, which makes it possible to take partial responsibility for the other’s
actions and doings as an extension of one’s own actions and doings.
96 C. COWLEY

will acquire, in 10 or 20 years. In addition, the “I” who is taking the

responsibility at the altar will also change more or less predictably, more or
less subject to my choices, in 10 or 20 years, and this will affect the oner-
ousness (on me) of my spouse’s unpredictable needs. So the marital vow
also involves something called “second-order” responsibility-taking: I am
taking responsibility for her future needs, as well as taking responsibility
for becoming the future person who will be placed in the situation of hav-
ing to meet her future needs. This then blurs my distinction between
“easy” and “hard” prospective responsibility-taking; instead, love requires
an ongoing re-taking of the responsibility.
Some will say, in an existentialist vein, that it is irresponsible to make the
unconditional vow in the first place, precisely because of all this ignorance
about future situations, needs, and selves. Far better to take one day, week,
or year at a time, governed not by a past vow made in ignorance, but by
an enduring policy of absolute honesty and good faith to other and to self.
If we drift apart, so be it. On the other hand, some would be inclined to
say that such “existentialist” love is not love; I’m not going to take a view
on that and will accept that deep and serious love is possible without long-
term commitment. Generally I am reluctant to compare intimate relation-
ships (“this one is more loving than that one”) because of the multiple
forms and layers of sheer particularity in each. What I do want to reject is
the existentialist’s claim that making the unconditional vow is necessarily
irresponsible. Taking responsibility for the unpredictable and onerous
future needs of one’s beloved is risky, certainly; but it need not be reckless.
There is a role here for luck, of course; if things go badly I might come to
judge that we had been reckless in making our vows, however informed
our commitment and good our prospects seemed at the time. In such
cases, Wolf’s quasi-­generosity might be relevant as a form of self-forgive-
ness in the face of ambiguous evidence to resist inappropriately destabiliz-
ing redefinition.
At the simplest level, taking responsibility at least means that one can,
as Solomon puts it, “promise to abstain from activities that will endanger
love. One can more positively promise to nurture conditions that are con-
ducive to love. One can even promise to adopt or strive for attitudes and
perspectives which are constitutive of love” (Solomon 2002: 26). This is
what I meant at the beginning by “indirect” choice being required for
love. But I would extend Solomon’s last sentence further in two ways.
First, it is not only a matter of adopting attitudes and perspectives, for such
adoption can be temporary, and such attitudes can remain “tacked on” in
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 97

a spirit of expedience in dealing with a crisis. What is more important is

not adoption but gradual transformation into a person whose perception
is naturally informed by the attitudes constitutive of love, and naturally
resistant to the centrifugal forces that can threaten any marriage. Second,
it is not just any love constituted by the attitudes I am adopting, it is love
for this particular person, and that requires a deep openness to the reality
of the other. And I argue that these two “extended” aspects of Solomon’s
account are more accurately called “taking responsibility for my spouse”
and that this is essential to marital love.
It might be tempting to say that taking responsibility in the aforemen-
tioned sense is supererogatory: that is, that it is admirable when someone
does it, but not blameworthy when a couple stumbles through a crisis
without being able to meet one another’s needs, or when the couple col-
lapses. Such heroism will be particularly apparent in cases of radical asym-
metry of needs, as when one partner has long-term health problems that
require sustained attention and effort from the other. Surely if anything is
supererogatory, long-term loving care is. And yet I would deny that this is
heroic, based on the subjective experience of the loving partner. From the
outside we might speak of admirable love, loyalty, and devotion; from the
inside, the loving partner merely does what she feels she “has to do.” This
leads to a paradox: I am calling “taking responsibility” a choice, and I am
saying that it is essential to love. And yet the loving partner, once she has
taken responsibility, finds herself in a position without choice, where she
has to respond to the needs of the other partner. And it is this lack of
choice, I contend, that undermines any attempt by third parties to label it
heroic. Or rather, a third party might still call it heroic, but the person will
not understand that as applying to him.7

5.5   Responsibility or Duty?

A lot of philosophical discussions have focused on the relationship between
love and duty. Insofar as one takes “responsibility” as a mere synonym for
duty, then the concept will add nothing new to that discussion. But I think

7
I stress that I am discussing this felt necessity within the context of a mutually respectful
adult relationship. If a wife suffers systematic abuse at the hands of her husband, but never-
theless declares that she “has to” take responsibility for him, I would not take this, by itself,
as an expression of love, and I would seek some larger description that allows for problematic
insecurity and dependence.
98 C. COWLEY

there are important differences between responsibility and duty, and that
teasing out the differences in this final section can tell us something useful
about taking responsibility. Recall the example of the one spouse who
decides to look after the other spouse with a long-term illness; we might
be inclined to call him heroic, but he would describe himself as doing what
“has to be done”—and I would call that taking responsibility. Could this
not also be called “acting on a sense of duty”? And does it matter?
Let me first summarize some of the issues between love and duty. One
famous conception of the contrast comes from Michael Stocker (1976),
who discusses a patient enduring a long recovery in a hospital. He is lonely,
bored, and restless. His friend comes to visit him. The patient thanks the
friend, and the friend replies: “not at all, I’m just doing my duty.” After
this, the patient gradually realizes to his horror that the friend is being
perfectly sincere. Stocker’s larger point is that, according to modern ethi-
cal theories of both Kantian and utilitarian flavors, the visitor is behaving
in an admirable way: he is sacrificing his own interests for the good of oth-
ers, motivated by a sense of selfless duty. However, in so doing he is not
being a very good friend, since the patient would reasonably hope that the
visitor would be motivated by love rather than duty; duty is essentially
impersonal, concludes Stocker, and merely seeks a contingent vehicle for
the visitor to maximize the good.
A contrasting conception of the contrast between love and duty comes
from Harry Frankfurt (1998). According to Frankfurt, love is essentially a
configuration of the will. For Frankfurt, love is about the necessities that I
encounter in my activities with the beloved: certain characteristic things I
find I “cannot” do, and others I “must” do, and Frankfurt takes these
locutions at face value. In terms of the agent’s moral psychology, con-
cludes Frankfurt, the necessities of love are just as real as the necessities of
duty. Sometimes this felt necessity will be surprising (pleasantly or unpleas-
antly) to the lover and will reveal the quality of the love in the sense that I
have been describing in the chapter.
My position differs from both Stocker’s and Frankfurt’s accounts, how-
ever. In response to Stocker, I would ask: what if it were very onerous for
the visitor to visit the patient on that day? Of course the patient would be
delighted if the visitor acted only out of affection, but circumstances might
conspire against such a visit, and then I would say that the patient should
be glad that the visitor comes, even if out of duty. Not only because it is
better than him not coming at all, but also because his motivating duty is
an expression of the respect with which he holds the patient. Okay, replies
 5 LOVE, CHOICE, AND TAKING RESPONSIBILITY 99

Stocker, but respect is not love; respect is compatible with affective dis-
tance and reluctance. We have to be careful here about definitions, but I
think a compromise position would be to describe the reluctant visitor as
acting out of a sense of responsibility rather than duty. Responsibility is
more than duty since it comprises a response to that particular patient and
his needs; it thus avoids the impersonality and stuffiness of duty that
rightly concerns Stocker.
The problem with Frankfurt’s account, on the other hand, is that it is
too unilateral. I can well accept the focus on felt necessity, but my worry
is that there is no essential reference to the partner’s actual needs, which
might first have to be ascertained with selfless attention. Frankfurt admits
that his account would allow for unrequited adult love. In contrast, I have
been looking at established relationships of mutual love and respect, where
the love is much more than an affectionate feeling, and it is shaped by the
response to the other and to her particular needs (or anticipated needs).
Such a response is a matter of perception, deliberation, feeling, and spon-
taneous action, and this is best captured in the concept of prospective
responsibility rather than necessity.

5.6   Conclusion
The philosophical literature on romantic and friendly love between adults
is quite sizable by now. There are distinct accounts about different mani-
festations of love, the varying phenomenology of love, the criteria for a
relationship to be deemed loving, the kind of self-understanding required
of a person loving another. I have argued that all these accounts share a
fundamental omission: while some speak of responsibility, it is never much
more than a corollary or consequence of love. In contrast, I have argued
that a robust notion of taking prospective responsibility belongs at the
center of love. For only such responsibility is sufficient to allow room for
the right kind of choice, and only such choice allows the love to be authen-
tic to the lover. Without such responsibility-taking, there is a risk that any
“love” will be rooted too much in capricious feeling or comfortable habit.
The fox did not speak of love, and it is not clear that the Little Prince
would have understood him if he had; instead the fox spoke of uniqueness,
of taming, and of responsibility.8

8
My thanks to Simon Cushing for comments on the first draft of this chapter.
100 C. COWLEY

References
Enoch, D. 2012. Being Responsible, Taking Responsibility, and Penumbral
Agency. In Luck, Value and Commitment; Themes from the Ethics of Bernard
Williams, ed. Ulrike Heuer and Gerald Lang, 95–132. Oxford: Oxford
University Press.
Frankfurt, H. 1998. Duty and Love. Philosophical Explorations 1 (1): 4–9.
Kolodny, N. 2003. Love as Valuing a Relationship. Philosophical Review 112
(2): 135–189.
Solomon, R. 2002. Reasons for Love. Journal for the Theory of Social Behavior 32
(1): 1–28.
St. Exupery, A. 2001/1943. The Little Prince (Trans. Katherine Woods).
Paris: Egmont.
Stocker, M. 1976. The Schizophrenia of Modern Ethical Theories. The Journal of
Philosophy 73 (14): 453–446.
Wolf, S. 2004. The Moral of Moral Luck. In Setting the Moral Compass. Essays by
Woman Philosophers, ed. C. Calhoun, 113–127. Oxford: Oxford University Press.
 CHAPTER 6

Not All’s Fair in Love and War: Toward

Just Love Theory

Andrew Sneddon

All’s fair in love and war. (proverb)

3. 1] To every thing there is a season, and a time to every purpose under

the heaven: 2] A time to be born, and a time to die; a time to plant,
and a time to pluck up that which is planted; 3] A time to kill, and a
time to heal; a time to break down, and a time to build up; 4] A time
to weep, and a time to laugh; a time to mourn, and a time to dance; 5]
A time to cast away stones, and a time to gather stones together; a time
to embrace, and a time to refrain from embracing; 6] A time to get,
and a time to lose; a time to keep, and a time to cast away; 7] A time to
rend, and a time to sew; a time to keep silence, and a time to speak; 8]
A time to love, and a time to hate; a time of war, and a time of peace.
(Ecclesiastes 3: 1–8; King James Version)

A. Sneddon (*)
Department of Philosophy, University of Ottawa, Ottawa, ON, Canada
e-mail: Andrew.Sneddon@uottowa.ca

© The Author(s), under exclusive license to Springer Nature 101

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_6
102 A. SNEDDON

6.1   Love and War: Getting Started

I take it that none of us believes that, in the name of love, anything goes.
The first documented appearance of the proverb about love and war in its
familiar form is in the novel The Relapse, or Myrtle Bank, which was pub-
lished in 1789, where it is cast as a “confounded lie”. The contrary idea,
that there are conditions under which the pursuit of love is appropriate,
and other conditions under which it is not, is ancient, as the famous pas-
sage from Ecclesiastes shows. These verses are notable for showing that the
consideration of the moral status of love and war in the same breath is also
ancient.
If one’s curiosity were sufficiently piqued by noticing the long-standing
juxtaposition of love and war, then one might become interested in the
details of their comparison. Is this linkage superficial, even coincidental?
Or is there wisdom here? To answer these questions, attention would need
to be given to similarities and differences between love and war. This
requires the mustering of some information; in advance of the work, we
can’t say whether there is anything of interest here.
But suppose that one is not particularly moved by the mere habit of
mentioning love and war together.1 Is there another reason to think that
there is some prima facie merit in studying love and war together? There
is, but to see it we need to attend to two ways in which we might conceive
of war. First, and quite naturally, we can think of war explicitly in terms of
conflict, and even (although secondarily) as characterized by animosity.
This way of thinking of war makes it inapt for functioning as a model for
love: not only does love not necessarily involve conflict or animosity, but
it is reasonably taken to be antithetical to both. Second, however, we can
think of war explicitly in terms of a common feature of human life with
very high stakes, especially the risk for harm. This approach offers a much
more promising model for love: besides also being a common feature of
human life, its potential for being the best part of a life, or the worst when
things go awry, is famous.
As it happens, the most famous body of thought about the ethics of war
conceives of it largely in terms of harm, and only secondarily in terms of
conflict and animosity. This long tradition of attention to questions
concerning the initiation and pursuit of war is known as “Just War

1
Indeed, some find the very idea of the exercise cynical. Hopefully the next two paragraphs
help to alleviate this worry. For more, see the discussion of Realism further in the chapter.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 103

Theory”. Seen at the right degree of abstraction, Just War theorists have
articulated a body of ideas concerning the achievement of the ends char-
acteristic of a common feature of human life while being sensitive to, and
minimizing, the harm that can be involved therein. At this level of abstrac-
tion, such ideas promise to apply straightforwardly to love. Whether this is
really the case, of course, depends on the details. Accordingly, here is my
plan: first, I shall briefly describe some of the ideas of Just War Theory.
This will demonstrate the harm-centric approach of this body of work.
Second, I will adapt the questions and concepts from Just War Theory to
apply to love. This will give us the framework for “Just Love Theory”. Just
as Just War Theory focuses on conduct and its stakes rather than on senti-
ment, Just Love Theory applies to conduct in the pursuit of loving rela-
tionships. Sketching some of the outlines of Just Love Theory will provide
some details to think about in order to see whether the ancient and peren-
nial juxtaposition of war and love is superficial or wise. I think that we find
something interesting and potentially fruitful here, but more details than
can be provided in this exploratory fashion are needed for a full evaluation.

6.2   Just War Theory

The concerns of Just War Theory are typically divided into three parts.
There are considerations governing when it is morally acceptable to go to
war, known as jus ad bellum considerations; considerations governing con-
duct during war, known as jus in bello; and considerations governing what
is owed after war has ended, known as jus post bellum issues. Undoubtedly
concerns relevant to love can also be divided this way, but I will focus on
conditions governing the beginning of a loving relationship and consider-
ations governing conduct during a loving relationship, so I shall put jus
post bellum concerns aside.
Here is an adapted version2 of what Helen Frowe offers as a standard
list of jus ad bellum conditions, that is, as criteria that must all be satisfied
for it to be morally acceptable to begin a war:

1. Just Cause;
2. Proportionality;
3. Reasonable Chance of Success;

2
I have dropped an independent condition of Right Intention as it is so closely linked to
the Just Cause condition, and since I will not be using it when I turn to love.
104 A. SNEDDON

4. Legitimate Authority;
5. Last Resort;
6. Public Declaration of War (Frowe 2011, p. 50).

Here are some details.3 The Just Cause condition is the most important
ad bellum consideration. Given that war is so destructive, it has been
thought that not just anything can be allowed to be a valid reason for
beginning a war. The centuries of thought about the Just Cause condition
have slowly settled on the idea that just wars must be a response. The gold
standard just cause for a war is an attack, such that the war is clearly under-
taken for defense. Trickier questions arise with the justifiability of first uses
of military force in response to threats.
More than a just cause is needed for it to be morally acceptable to enter
a war. The other ad bellum considerations are articulations of what else
besides a just cause is needed for there to be an overall acceptable case for
beginning a war. The Proportionality condition requires that military
force be a proportionate response to the just cause. War can be an over-
reaction to certain violations of national sovereignty. The Reasonable
Chance of Success condition gives voice to the idea that futile wars cannot
be justifiably begun. The condition of Legitimate Authority is used to
distinguish between such conflicts as inter-state wars on the one hand and
unorganized conflicts between mobs on the other. Violent conflicts involv-
ing non-state actors, such as terrorist organizations, fall between these
poles and raise tricky questions as to whether they do or even can count as
“wars”. The Last Resort condition requires that less destructive options
be tried to address the just cause before war is undertaken. The Public
Declaration condition is self-explanatory; its rationales include allowing
public debate about the war, last-minute changes of heart by the target of
the declaration, and efforts aimed at enhancing the safety of civilians.
Jus in bello considerations can be articulated in various ways.4 I shall
divide the territory into three conditions that must be met for morally
acceptable conduct during war:

3
Just some, to give a sense of the territory. Frowe notes that there is wide agreement that
these are the formal conditions for beginning a war, while there is significant dispute about
the substantial details of each condition. This is not the place to delve into these disputes.
4
For example, contrast Frowe (2011, Chap. 5) with Lazar (2020).
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 105

A] Necessity;
B] Proportionality;
C] Discrimination.

The Necessity condition holds that attacks during war must be of mili-
tary necessity. This means that they cannot be for such things as pillaging
or revenge or ethnic cleansing; they must, somehow, serve the just cause
of the war. The Proportionality condition requires that the destructive
means of waging war be constrained by the military ends that particular
operations aim to achieve. Widespread death and destruction merely to
attain a slightly better physical position on a battlefield is ruled out by the
values at work here, for instance. Finally, the Discrimination condition is
meant to make precise the suspicion that not all possible targets of military
action are legitimate. Those waging war must distinguish between people
directly involved in the violent conflict—the combatants—and people
who are not. Non-combatants cannot be legitimately targeted or attacked.
Just how much they can legitimately be put at risk is a tricky issue.
Overall, the idea driving both the ad bellum and in bello parts of Just
War Theory is that war is so destructive that it needs to be analyzed and
conducted very carefully if it is going to be morally acceptable. In light of
this idea, Frowe adds another in bello consideration: Realism. The so-­
called realist contends that the chaos and fear experienced by soldiers dur-
ing war makes the attempt to articulate conditions for its ethical governance
and control naïve. War is hell, and there are no ethical rules in hell. There
are various ways that the Realist position can be articulated. One way is
put in terms of whether ethical concerns apply to war at all, and as such
violence is an inevitable feature of human nature. Seen this way, the
Realism condition is not so much in bello as about war at every level of
analysis, with the realists contending that war is outside of the domain
of ethics.

6.3   Toward Just Love Theory

Realism
Let’s apply these distinctions to love, starting with Realism. Why should
we think that love is subject to ethical analysis and regulation? After all,
love is something that (in)famously befalls us. We fall in love; the meta-
phor suggests the inescapable pull of gravity at work with regard to who
106 A. SNEDDON

and when we love, rather than intentional action and an agent’s control.
For there to be anything like a Just War Theory for love, love must be
shown to be, in general, the kind of thing that it makes sense to evaluate
ethically.5
While I’m sure that there’s much interesting things to be said about the
aptness of the metaphor of “falling” with regard to love, I’m also inclined
to grant this much of the objection.6 I shall assume that love, the experi-
ence, happens to us. But this leaves much for the ethicist to discuss. Let’s
suppose that love (the sentiment) is in place. What’s to be done in the
light of this? While in the grip of love, what may one do? What may one
not do? Of particular importance are actions that concern the object of
one’s love. Just as famous as the fact that we fall in love is that we tend to
want to do things when in love: to make declarations of love to the
beloved, to pursue special sorts of relationship with loved ones, to spend
time with these people, to perform actions of a wide variety of kinds, some
that we are very unlikely to perform or even to want to perform with
people we do not love. In short, love is one thing, being in a loving rela-
tionship is another. Ethical considerations straightforwardly apply to lov-
ing relationships. Following the distinction between jus ad bellum and jus
in bello considerations, we can ask questions about the conditions under
which it is ethically permissible to begin a loving relationship and other
questions about the sorts of conduct that are permissible within such a
relationship. Following the Latin naming convention found in Just War
Theory, I shall call the ethical criteria for acceptable beginning of a loving
relationship “jus ad amantes necessitudo” conditions. The criteria for
acceptable conduct within a loving relationship are here called “jus in
amantes necessitudo” conditions.
Just what counts as a loving relationship? Heterogeneity rules here;
consider the similarities and differences among relationships between
(respectively) close family members, lovers, neighbors, friends, and more,
as well as variety within these categories. A full Just Love Theory would
attend to the relevant details, but space prohibits such an extensive inquiry

5
With regard to war, the Realism concern sometimes takes the form of a worry that trying
to do ethical analysis in this domain is naïve. With regard to love, the analogous worry would
be that such analysis is cynical: love is inherently good, so raising questions about its justifica-
tion must be to diminish this. I won’t address this directly, other than to counsel patience
with the details of the discussion in order to assess whether this sort of inquiry is necessarily
cynical.
6
For discussion of the justification of the experience of love, see Sect. 6 of Helm (2017).
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 107

here. I shall confine my attention to the relationships found between, col-

loquially speaking, lovers.7 This territory can only be characterized impre-
cisely, but enough can be said for present purposes.
I propose to use the experience of love as a standpoint for describing
what counts as a loving relationship between lovers. My assumption is that
the sentiment provides a regulative ideal for understanding loving rela-
tionships: they are the relationships called for by the experience of this
emotion. Detailed analysis of the phenomenology of love would be help-
ful, but space prohibits such work. Instead, here are what I take to be all
but platitudes about love and loving relationships:
First, the experience of love is other-directed:8 we fall in love with (at
least) another person. Accordingly, a loving relationship (of the sub-­
species being currently examined) is a relationship with at least one
other person.
Second, the sentiment of love is intimately tied up with other emotions.
Happiness and sadness are obvious ones; jealousy is hardly less obvious;
longing, worry, wistfulness, and, undoubtedly, others are still only just a
little bit less obvious. I take this as indicating that loving relationships are
sources of various kinds of emotional satisfaction and disappointment.
Third, somewhat more controversially, I take it that our emotional
experiences are, to some significant extent that I cannot make precise,
experienced holistically when in love. Whereas happiness about an achieve-
ment when one is not in love might be experienced just as this happiness
alone, such happiness is often (if not necessarily) experienced as something
to be shared when one is in love. It is entwined (to some significant extent)
with one’s affection for another, with the other’s feelings about this proj-
ect and your own feelings, and more. Accordingly, loving relationships are
ones which tend to give rise to such holistic emotional experiences. Indeed,
this is one way in which we might make precise the idea that in loving
another, one opens oneself up for transformation. We might be able to

7
There are other linguistic clues regarding the kind of loving relationship that is the pres-
ent topic. It is the sort of relationship we have with people with whom we are “in love”. It is
the kind of relationship where it is appropriate to speak of one’s “beloved”. In English,
“lover”, “in love”, “beloved” (and maybe other terms) are not used to characterize the senti-
ment, relationship, person, or other sort of thing loved as kin, friend, neighbor, country,
favorite food, etc. Philosophers have long distinguished different forms of love: eros, philia,
agape. The present topic falls within eros. See Helm (2017) and Moseley (2020).
8
This does not preclude it also being self-directed, as “union” theories of love claim: for
example, Scruton (1986), Solomon (1988), Fisher (1990).
108 A. SNEDDON

take the extent to which a relationship has this sort of holistic, identity-
transforming effect on one’s emotional life as a measure of its centrality to
a person, or, in other words, as a measure of the depth and significance of
the love.
Finally, and relatedly, the experience of love is not a passive stance taken
toward another person. Typically, it gives rise to desires to do things with
the other person. None of these activities is necessarily unique to love.
Famously, love is experienced as, in part, a desire to be with the other
person, and perhaps to live one’s life with that person. The activities
involved are both famous and, when listed, banal: talking, eating, walking,
sex, and more. The more one wants to do all of the activities of ordinary
life with a specific other person, the more one wants to be in a loving rela-
tionship with that person. More significantly, loving relationships are not
isolated from the rest of our lives: indeed, our very experience of love
implicitly acknowledges the value of the myriad activities of living, as
going through these activities with the beloved to a significant extent
makes real, or perhaps completes, the loving sentiment one has to the
beloved. Again, this is a way in which being in love can be transforming.
This is not an exhaustive list. I take these characteristics to be typical of,
but not necessary or sufficient for, loving relationships. This gives us the
somewhat odd but important implication that one can be in a loving rela-
tionship without either party having the sentiment of love. For example, if
the parties in question pursue their lives together, want to spend time
together, but lack the emotional holism that I have noted, then we should
think that they are in a loving relationship without the sentiment of love.
I take it that such total absence of love-as-an-experience will be quite rare.
Still, this conceptual space makes sense when we stand back and think
about the varieties in which the experience of love comes—heated,
“puppy”, mature, deep, shallow, whole-hearted, and more. Loving rela-
tionships are possible for this whole array, requiring some looseness
between the experience of love and the nature of loving relationships. We
shall see that this is important for taking stock of the goals that might be
sought through loving relationships.
These reflections about loving relationships sharpen what can be said in
response to the Realism worry. Consider war: it is important to enter and
to wage carefully, in some ways and not in others, because of the death and
destruction that it causes. It is a massive source of waste, which can only
be justified on particular grounds. The same goes for the activities of love,
albeit these must be described on an individual scale rather than from the
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 109

perspective of such groups as nations. As with war, the risk of waste in the
pursuit of love calls out for ethical management. To enter and to conduct
a loving relationship demands large amounts of resources from us, of vari-
ous kinds. Somewhat superficially, it consumes our wealth. More deeply,
to commit oneself to another in a loving relationship uses one’s time,
one’s emotions, one’s attention, and generally what one has to give to
others. When we enter a particular loving relationship, we can’t use these
resources for other things, including a vast array of other particular loving
relationships. The opportunity costs of love are massive. Without care, this
all amounts to mere waste.9
It is natural to focus on one’s own perspective when thinking about
entering and conducting a loving relationship, such that the risk of waste
can seem thoroughly self-regarding. I think that this suffices to make this
an issue of ethical import, but others will disagree, finding the idea of self-­
regarding duties suspicious. This doesn’t matter, as the self-regarding per-
spective is not all that matters here. The risk of waste concerns other
people too, by the very standards of the sentiment of love. Love is for
others and involves desires to do things with those others: other people
are necessarily drawn into, and shut out of, the sort of relationships that
are our topic. This centrally includes the beloved, but third parties are
affected here too. Since the effects of our actions on other people are
uncontroversially of ethical significance, we should think that the activities
involved in loving relationships are straightforwardly subject to moral
standards.

Jus ad Amantes Necessitudo

The jus ad bellum conditions concern when, if ever, it is justifiable to begin
a war. Analogously, jus ad amantes necessitudo considerations pertain to
the question, “When, if ever, is it justifiable to begin a loving relation-
ship?” I will not weigh all of the ad bellum considerations equally in con-
structing ad amantes necessitudo counterparts. The least important is the
condition of “Legitimate Authority”. For love, the issue is whether the

9
Strictly speaking, age makes a difference to the line of thought in this paragraph. Simply
put, the older one is, the less one is putting at risk for oneself in entering a loving relation-
ship. The same goes for the other person brought into the relationship, and for third parties.
It’s an oversimplification to say that there is an inverse relationship between age and stakes,
but it’s not much of one.
110 A. SNEDDON

person who seeks to establish a loving relationship can legitimately do so.

Since love is the relevant affection, the person seeking the relationship
must be capable of loving. Since a relationship with another person is
sought, the person seeking the relationship must be capable of conducting
such relationships. Barring particular reasons that undermine either of
these capacities, psychologically competent people should be thought to
be “legitimate authorities” with regard to seeking loving relationships.
Among the sorts of things that undermine such legitimacy are (a) psycho-
pathologies that preclude either love or being in personal relationships,
and (b) situational contingencies that put such love or relationships out of
reach. Let’s call this the condition of “Psychological Competency”.
The first significant jus ad bellum condition for present purposes is “just
cause”, which has traditionally been glossed as “a wrong received”: it is
justifiable to begin a war only as a response to certain sorts of damage or
threat. There is no reason to think that loving relationships need to be
started only as responses, especially to wrongs. Loving relationships are
different from war in this respect. When we seek to begin a loving rela-
tionship, we aim to establish a particular sort of relationship with another
person, and we aim to do particular things with the other in virtue of love
for them. Accordingly, instead of “just cause”, I suggest that to begin a
loving relationship justifiably, one must be aiming at a “just target” with
whom one proposes to pursue “just goals”.

Just Target
What is it for a person to be a “just target” for enticing into a loving rela-
tionship? Broadly and roughly speaking, the person at whom one is aiming
must be “available” for joining in the relationship. “Available” here does
not mean “physically local”: I don’t see why we should not pursue loving
relationships that are conducted over long distances. Trivially it means
“alive”, but the moral problems associated with wasting our time trying to
enter relationships with persons who don’t exist (due to death or due to
never having existed) strike me as worth putting aside. Instead, I mean
available from the standpoint of the sentiment of love, which desires to
establish a loving relationship with another person.
There are three specific components to such availability. First and sec-
ond, the person in question must be available in the sense of being able to
join the relationship. As with the Psychological Competency condition,
two things must be in place: the person must be capable of love, and they
must be capable of participating in close personal relationships. All but
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 111

rough assessment of these capacities in another is impossible for the vast

majority of people. Where general psychological competency is in place,
and where there are no specific reasons calling into question these two
capacities, these aspects of the just target condition can be assumed to
be met.10
The third component stems from the stakes involved in joining a loving
relationship. When one tries to begin a loving relationship with another
person, one asks that person for a lot. To join the relationship, the other
must spend time, money (in all likelihood), and psychological resources
on you and the relationship. This is at least costly, and potentially risky. Let
me suggest that it is prima facie morally problematic to impose such costs
on others without their agreement. If this is granted, then a constraint is
placed on loving relationships: we must only try to begin these with peo-
ple who are capable of consenting to the costs that they involve. Consent
is here understood formally: the target of one’s love must possess the
capacity to consent to joining in the relationship. Familiar ethical prob-
lems arise when we seek to perform complex and potentially damaging
activities with and on those who are incapable of consenting to being
included in these activities. As we shall see, this does not exhaust the sig-
nificance of consent to the beginning of a loving relationship.
The jus ad bellum Just Cause condition conceives of war as sometimes
an appropriate response, typically to a wrong received. The cause is worth
responding to with violent defense. Analogously, we can ask whether, to
be a just target of inclusion in a loving relationship, one must be worthy
of love. We should distinguish two dimensions of such worthiness:

Evaluative
One is the sort of person whom it is good to have in such relationships.
Seen this way, the worthiness condition is satisfied in general by anyone
who meets the capacity conditions of being a just target: generically, this is
the sort of person whom it is good to love. Specifically, we can tell whether
someone is good to love only by attending to particular details about per-
sonality. We all know that these vary interpersonally: a person good to

10
It is important not to insist on psychological normalcy. We should not assume that
abnormality implies incompetence; one can be psychologically abnormal yet competent, and
it is competence that matters. Nor does normality guarantee competence: it is normal for the
very young not to be psychologically competent.
112 A. SNEDDON

include in a loving relationship with A might be unsuitable for a loving

relationship with B.

Deontic
One is the sort of person who is owed such a relationship. I think that we
should dismiss this way of understanding individual worthiness in relation
to loving relationships. The reason is basic: the beginning of such a rela-
tionship is supererogatory, and hence never owed to anyone. We are all
greatly fortunate when others include us in loving relationships (ceteris
paribus), but we are not wronged when we are not included.
The effect of analyzing worthiness of love along these dimensions is to
make clear that this is not a particularly important aspect of being a just
target for bringing into a loving relationship.

Just Goals
When we seek a loving relationship with another, “loving relationship” is
probably not what we have in mind (although it might be). Rather, we
have instead (or as well) more particular goals. We can divide these goals
into (1) those that are internal to a loving relationship and (2) those that
are external.

Internal Goals
By goals “internal” to a loving relationship, I mean those that have to do
intrinsically with being in or running a relationship. Being partners with
another is one such goal: “partner” is the name of a particular kind of
relationship. Being in control of the other person is another internal goal.
It is possible that personal transformation is an internal goal. If I aim to
transform myself by leading my life deeply involved with another person,
then such transformation must be pursued within a loving relationship.
But if I aim instead at changing myself, or even at just being open to such
change, then this can, in principle, be sought without being in a loving
relationship.

External Goals
Goals that are “external” to a loving relationship are those that can be
pursued without being in such a relationship. Sex, having children, secur-
ing care in one’s old age, company, and, perhaps, intimacy are external
goals in this sense.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 113

These characterizations are not meant to be exhaustive: I have not

attempted to list all possible internal and, especially, external goals. More
significantly, there is no reason to think that one must have only one sort
of goal in mind when one seeks to being a loving relationship, although
one might. Human motivation is sufficiently complex to include both
kinds. Most importantly, whether a goal of either kind is “just” is a com-
plex matter. In some cases, goals will have features that count prima facie
for or against them. I take partnership to be prima facie acceptable and
control to be prima facie suspicious. Some goals have both favoring and
disfavoring features. Goals that involve using the other can be like this: use
of the other is prima facie morally problematic, but the particular purpose
of the use might be prima facie acceptable or even laudable. To a certain
extent, the overall justifiability of the goals sought in establishing a loving
relationship is not independent of the other ad amantes necessitudo con-
siderations, so let’s turn to them.
I shall start with “Public Declaration”. Regarding violent conflict, for
discursive and risk-minimization reasons, war must be publicly declared to
be legitimate. For starting loving relationships, public declaration of one’s
intentions, especially regarding the goals that one seeks via a loving rela-
tionship, is significant because of the importance of consent.11 Accordingly,
let’s call this the “Public Declaration and Consent” condition. Whereas
the general capacity for consent is part of what makes you a just target for
inclusion in a loving relationship, this does not suffice to make any loving
relationship that includes you morally justifiable. For this, you must actu-
ally have consented to the relationship. Consent thus matters to jus ad
amantes necessitudo both formally and substantially. Substantially, the tar-
get of one’s love must be given a real chance of consenting to being in the
relationship. For this, open declaration of one’s intentions is needed. This
requirement rules out deception of the would-be beloved by the person
seeking a loving relationship. For consent to be genuine and authoritative,
the person giving consent must have a clear view of that to which they are
consenting. Deception obscures such a view. When I lie or obfuscate about
my intentions, my situation, my prospects, my health, or anything else
relevant to beginning a loving relationship (which, in principle, is anything
whatsoever), then you cannot completely consent to being in this sort of
relationship with me, as details relevant to your consent are kept from you.

11
Kyla Ebels-Duggan (2008) makes much of the significance of the active participation of
the beloved in love.
114 A. SNEDDON

The more significant these details are to the nature of the particular rela-
tionship, the more such deception compromises the legitimacy of my
bringing you into this relationship.12 Public declarations must be answered
in order for a loving relationship to be begun. Consent must actually be
given; legitimate loving relationships of the sort here being examined can-
not be unilaterally established.
We should not assume that public declarations of intentions can take
just one form. For one thing, they can vary in explicitness. Morally, the
more explicit the declaration, the better it is, but suboptimal ways of shar-
ing information can still satisfy this condition. Also, we should not insist
that there be just one such declaration happening before the very begin-
ning of a loving relationship. Relationships change over time, as do goals,
so there can be good reason for more-or-less outright public declarations
within a relationship that has already begun. Wanting something new can
require a substantially new relationship with the same person. Consider a
change from a relationship that has only internal goals to one that has
external ones as well. Still, insisting that this requires a new relationship
seems unrealistic to me, making the Public Declaration and Consent con-
dition both ad amantes necessitudo and in amantes necessitudo.
So far, the start of loving relationships and the significance of public
declarations have been considered from the standpoint of a person seeking
such a loving relationship with another and hence of the person making
such a declaration. It is worthwhile to consider the perspective of the party
who receives the declaration. Does the fact that another person has made
a declaration to you put you in a different ethical situation from that per-
son vis-à-vis the beginning of a loving relationship? So far as I can see, it
does not: even when another has approached me, I can only enter a loving
relationship if that person is a just target, and I am as much on the ethical
hook for declaring things about me as the other person is. In what follows,
I shall assume that there is no ethical difference in the standpoints of first
declarers and responders with regard to the beginning of and conduct
within loving relationships.

12
The significance of substantial consent also rules out coercing one’s target into the rela-
tionship. While one might reasonably doubt whether a loving relationship can really be
established through coercion, I suspect that human activities and feelings are sufficiently
complicated to include this as a real possibility, and this makes its prohibition from the stand-
point of Just Love Theory significant. Systemic power imbalances can function as coercive
forces that undermine the legitimacy of loving relationships sought by the powerful with the
powerless.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 115

Consent clearly matters to the moral significance of the goals sought via
a loving relationship. If I am trying to get you into a loving relationship to
use you in a certain way, your consent to the use matters centrally. Consider
seeking a relationship in order to have children: consent to the relation-
ship for this purpose will be very important if it is to be morally acceptable.
Abstracting from particular examples, the general ad amantes necessitudo
significance of consent is best understood not in terms of consent just to
the relationship abstractly understood, but rather in terms of consent to
the goals sought through establishment of the relationship.
Consent is not all that matters to the moral status of the goals sought
via beginning a loving relationship. The other ad amantes necessitudo con-
ditions also matter. I shall address these first in connection with goals
internal to loving relationships, then in connection with external goals.
The jus ad bellum condition of Last Resort requires that other options
be tried before the destructive endeavor that is war is embarked upon.
This in part captures the idea that, if it is to be justified, war must be neces-
sary, and to ensure that this is so less destructive options must be tried
first. Since entering a loving relationship is significantly risky and costly,
both for oneself and for others, it is worth considering whether it too is
really necessary. Let’s rename the relevant jus ad amantes necessitudo con-
dition “Necessity”. For goals internal to loving relationships, the answer
to this question is an automatic “yes” since they cannot be achieved out-
side of such relationships.
The jus ad bellum Proportionality condition requires that war be a pro-
portional response to the just cause, rather than a massively destructive
overreaction. Again, since significant stakes for multiple parties are
involved when a loving relationship is started, it is well worth considering
whether the relationship in question is a proportional step to take regard-
ing what one wants to achieve. It too might be far more than is justified
by one’s goals. For goals internal to loving relationships, this criterion is
all but automatically met. After all, the sentiment of love calls out pretty
automatically for trying to establish a loving relationship with the beloved.
What makes it the case that this condition is not automatically met is con-
tingent details. These details are particularly important regarding one’s
sentiments about the putatively beloved and in relation to one’s goals.
Where one is mistaken, unclear, or self-deluded about what one wants or
whether one really loves the person one is trying to get into a loving rela-
tionship, problems of proportionality can arise. To the extent that one
already genuinely loves the target of one’s efforts, a loving relationship is
116 A. SNEDDON

a proportional measure. Interestingly, this means that attempting to estab-

lish such a relationship with a stranger can be acceptable by the standards
of the Proportionality condition. Consider infatuation with a celebrity: a
person can, in principle, know a lot about such a person and genuinely
love them without ever having met them. Trying to get them into a loving
relationship is not a disproportionate response to one’s sentiments in such
a case, but it will likely fail by other jus ad amantes necessitudo
conditions.13
The jus ad bellum condition of Chance of Victory says that futile wars
are unjustifiable: their destruction is all waste. The “chance” involved here
is neither a matter of guarantee nor of bare possibility, but of reasonable
success. Once again, since entering loving relationships is risky, indeed,
since even bringing up the possibility of such a relationship with someone
is risky, then doing so ought to be undertaken only when there is some
significant chance of success. Accordingly, let’s rename the relevant jus ad
amantes necessitudo condition “Chance of Success”. Failure here is all but
guaranteed when one’s public declaration is turned down.14 Other factors
count against success to a lesser extent. An obvious case is when the
beloved is already in a loving relationship with another. A slightly less clear
case is when the target of one’s affections loves another but is not in a lov-
ing relationship with them. A difficult case concerns a public declaration
to which the response is hesitant: the other person is unsure and wants to
keep things non-committed for the time being. In such a case, continuing
in hopes of establishing a loving relationship, or as if such a relationship is
already in place, is deeply risky and arguably unjustified.
So much for goals internal to loving relationships; let’s turn to external
goals. The jus ad amantes Necessity condition asks whether a loving rela-
tionship is necessary for the goals in question. For external goals, the
answer is an all but automatic “no”: by definition, goals external to loving
relationships can be achieved without these relationships. This holds in
principle and in a vast array of actual cases. However, it is possible for cir-
cumstances to be such that some external goal cannot be achieved outside

13
There is debate about whether love should be construed as a response to the value of the
beloved: e.g., David Velleman (1999, 2008). Such a view opens up the possibility that an
attempt to start a loving relationship with another can be a disproportionate response to the
other’s value. See Helm (2017) for critical discussion of Velleman’s view.
14
So far as I can tell, success is impossible where the target does not exist, but I will refrain
from insisting on this. People have powerful imaginations and rich lives in virtue of this,
after all.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 117

of a loving relationship with a particular party. In such rare cases, the

Necessity condition is satisfied by external goals alone.
Given this point, the Proportionality condition will also typically be
violated when one seeks to enter a loving relationship purely for external
goals. Since these ends can usually be achieved without such relationships,
to enter a loving relationship is to put unreasonably much at risk for them.
It’s not just material resources that matter here; it’s our time and affec-
tions at least as much, and I, for one, am inclined to value these sufficiently
highly as to count it unjustifiably risky to begin loving relationships solely
for such external goals as sex, children, care in old age, and more.
The Chance of Success condition rules out some loving relationships
sought solely for external goals when these goals cannot realistically be
achieved, either at all or through this sort of relationship. A heterosexual
who seeks a loving relationship with a homosexual for the external goal of
sex will fail by the standards of this condition. So too will a person who
seeks a loving relationship for intimacy from someone who does not like
them. More ambiguously, degree of consent to a declaration can make a
difference here. Where one consents to some aspects of what is declared
but not to others, the chance of the relationship succeeding (allowing
varying metrics of success here) is affected, and in some cases it might
be doomed.
The upshot of these considerations is that it is very difficult for external
goals alone to make entering a loving relationship justifiable. It is possible,
but it is not at all likely. My guess is that, in reality, this does not matter
much, as most people have internal goals in mind as well as external ones.
However, this is an empirical wager, and I could easily be wrong. Ironically,
one of the risks of entering a loving relationship solely for external ends is
that one party will come to want internal goals as well, to the detriment of
the relationship. The risks of love are complicated.

Jus in Amantes Necessitudo

The jus in bello conditions pertain to what sorts of things can justifiably be
done during the course of war. The “Necessity” condition limits the justi-
fiable use of the violent power of war to those ends that are of military
necessity. The benefits and, especially, harms that are at stake outside of
military parameters still matter when war is going on; the Necessity condi-
tion enjoins soldiers and military decision-makers to give them their due.
118 A. SNEDDON

The same goes for conduct within a loving relationship: it takes place
within lives that are already up and running and for which there are stakes
independent of the loving relationship. While the sources of benefit and
harm that exist prior to a loving relationship might be transformed by it,
they are nonetheless to a significant extent independent of it. Such things
as health, financial stability, and relationships with third parties still matter
after a person becomes immersed in a deep loving relationship. They mat-
ter prudentially, of course: if I am in a loving relationship, impoverishing
myself for love leaves me worse off than I might be because I can’t pay my
bills or do other worthwhile things with my wealth. But they also matter
interpersonally, and hence in a classically morally relevant sense. If lovers
encourage each other to do physically risky things because they are such
enjoyable ways of pursuing their loving relationship, then they might well
end up with health problems that leave them unable to care for each other
at some later time. Consent to these activities mitigates the problem within
the relationship, but poverty and ill-health also affect our abilities to do
things for third parties. Our relationships to third parties take a wide vari-
ety of forms and involve lots of sorts of duty. A parent who is unable to
take care of children properly because of excessive spending on a partner
in love has done something morally remiss. Accordingly, we ought to keep
the pursuit of love in check: loving relationship is deeply worthwhile and
hence worth encouraging and pursuing, but the costs/risks involved
should be measured by reference to thought about just what is necessary
to the existence and well-being of the relationship. Beyond this, we risk
wasting resources, thereby diminishing the quality of our lives and failing
in our moral obligations.
Let’s note three points of detail about the jus in amantes necessitudo
territory of Necessity. While I have put the issue in terms of the indepen-
dent value of aspects of our lives other than loving relationships, this is not
merely a recognition of the general point that various things matter to us.
It is a point internal to the nature of love and loving relationships them-
selves. Recall the fourth all-but-platitude about love and loving relation-
ships presented earlier: to love another and to have a loving relationship
with that person is to a significant extent to want to run one’s life with
them. So it is a failure by the standards of love itself to lose sight of the
independent value of other aspects of human life. Although I won’t lean
on this point too heavily, what this means is that involvement in a loving
relationship that neglects the Necessity condition will tend to be, arguably,
self-defeating: a loving relationship that consumes the rest of one’s life at
least risks undermining itself.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 119

Second, there is no reason to think that it should be easy to be able to

identify whether some activity or expenditure is necessary either to the
existence or to the flourishing of a loving relationship. How could it be?
This is to a significant extent a matter of knowing what it is for a life to go
well, and such practical wisdom is not had by all. The lack of a measuring
procedure here should not lead us to think that this is impossible, how-
ever: lots of people manage to pursue both loving relationships and other
things. Relatedly, the third point is that this is not an all-or-nothing mat-
ter. Sometimes we will give too much time to a loving partner, sometimes
not enough. Since jus in amantes necessitudo conditions pertain to con-
duct during a loving relationship, they cannot be satisfied all at once, for-
ever. Practical wisdom matters day-to-day.
The general spirit of the Necessity condition overlaps with that of
“Proportionality”. The jus in bello Proportionality condition holds that
the use of destructive military powers must be proportional to the particu-
lar ends to be achieved. Likewise for love: there are many things we might
do within a loving relationship, and hence many costs that we might incur.
Some of these will be disproportionate to the value of the end to be
achieved. To give up a kingdom for a kiss is too much (arguably). More
realistically, let’s say that loving relationships generally require lovers to
pay attention to each other and to spend time with each other. Too much
attention is obsession; too much time spent on another, such as when she
is trying to pay attention to other parts of her life, is stalking. That we
already discourage these degrees of expenditure in the name of love indi-
cates extant sensitivity to jus in amantes necessitudo concerns of
proportionality.
Proportionality has both self-regarding and other-regarding aspects.
Part of the idea is that, for one’s own good, one should not overdo it with
regard to the pursuit of love. The other part is that the costs/risks that
come with loving relationships to which others are exposed should be kept
to a minimum. “Others” includes one’s beloved. Interesting details are
revealed when we focus on the other-regarding aspects of conduct during
love. Let’s assume that all such conduct involves costs (time, emotion,
wealth, and more). Let’s assume also that all such conduct is performed
for the good of the relationship. The actions in question can be further
subdivided: some will be good independently of how they conduce to the
relationship, whereas others will be intrinsically morally problematic. Since
we ought to minimize the harms imposed on the beloved in the course of
love, we ought, ceteris paribus, to choose actions that are intrinsically good
120 A. SNEDDON

if possible, and not ones that are intrinsically problematic and only instru-
mentally good. Suppose that I want to deepen the commitment of my
lover to me and our relationship. I might try to do this by slandering oth-
ers to whom my beloved is emotionally attached, thereby weakening her
other emotional ties and close relationships. Alternatively, I might try to
make her fear life without me. I take it that to cause fear in another, espe-
cially one’s beloved, and to weaken emotional ties and relationships are
prima facie problematic. Such actions should be eschewed in favor of ways
of deepening the relationship that are themselves morally laudable.15
It’s not just our partners in love who are affected by the conduct we
pursue within loving relationships. There are third parties also. The jus in
bello condition of “Discrimination” distinguishes combatants from non-­
combatants and prohibits at least the targeting of non-combatants in mili-
tary operations. The general idea is that the harm of war should be limited
to those directly involved in it. Much the same holds jus in amantes neces-
situdo: we should not deliberately subject third parties to the risks/costs
involved in the pursuit of our loving relationships. Consider again my
slander in the previous paragraph: to the extent that this harms not just my
beloved but also the person slandered, I ought not to do it, even in the
name of love. There are more subtle issues here too. First, besides acts that
cause harm, there can be harm due to inaction. Third parties who depend
on us emotionally or materially should not be harmed through neglect
due to diversion of our attentions by love. Second, besides the intrinsically
morally problematic acts that might be performed to foster a loving rela-
tionship, the relationship itself (the love and the conduct therein) can be a
burden on third parties. It can be an emotional burden, as when people so
wrapped up in their love make others uncomfortable through displays of
affection, inside jokes, pet names, idiosyncratic patterns of talk, and so
forth. But it can also be a material burden, as when lovers deliberately rely
on third parties to do things that the lovers should be doing but are not
because of their relationship. I once did an unduly large part of the late-­
night cleanup jobs in a fast-food restaurant because a co-worker was patch-
ing up an argument with a boyfriend. In retrospect, this was irresponsibly
much of them to ask of me.
For a third thing, third parties can be involved in loving relationships in
complex ways. This relates to some of the practical and conceptual

15
Presumably actions can be good in some respects, bad in others. Action-individuation is
a famously vexed topic, so I won’t dwell on the issues here. See Ginet (1990) and Mele (1997).
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 121

difficulties probed in the jus in bello context. It is wrong to target non-­

combatants; what about putting them at risk? What about incidentally
harming them as so-called collateral damage? Do the values here prohibit
deliberate use of non-combatants or objective harm of them? It is likewise
for love. Consider a loving relationship with someone who loves other
people: parents, siblings, children, and so forth. One might have no inter-
est in love for or a loving relationship with any of these people directly, but
what about as part of the loving relationship with one’s lover? Is it accept-
able to direct loving activities at third parties as the means of achieving
things in one’s loving relationship? What about creating people—having
children—purely to serve a loving relationship? It might not be strictly
possible to love any of these people as means; arguably this requires loving
the person in themselves, directly (Stocker 1976). Still, activities charac-
teristic of love and of certain sorts of loving relationships can be per-
formed. Communicative aspects of acts are particularly important here:
declarations of affection to third parties as means of pursuing a loving
relationship with one’s beloved put the third parties at emotional risk at
least. I’m inclined to think that the details here are sufficiently complex as
to preclude saying that such involvement of third parties in the conduct of
loving relationships is outright prohibited, but this complexity should also
give us pause. Some ad amantes necessitudo conditions will apply to third
parties as well. We bear a burden of caution at least regarding the ways and
extent to which we draw third parties into loving relationships.

6.4   Conclusion
Here is a list of the contours of the aforementioned Just Love Theory:

• The Realism Condition

• Jus ad Amantes Necessitudo:
–– Psychological Competency
–– Just Target
–– Just Goals
–– Public Declaration and Consent
–– Necessity
–– Proportionality
–– Chance of Success
122 A. SNEDDON

• Jus in Amantes Necessitudo:

–– Necessity
–– Proportionality
–– Discrimination

This is not necessarily exhaustive. The moral contours of war need not
map perfectly onto those of love. For a complete Just Love Theory, other
sorts of love need attention. Still, there is much here to constitute the start
of such a theory.
One thing still missing is consideration of ethical issues pertaining to
the end of loving relationships. Here love and war tend to differ deeply.
There is a natural end to war, dictated by the nature of its just cause: once
that has been addressed or achieved, the war should end. When loving
relationships are sought solely for goals external to the relationship, there
can be such a natural end. However, I take it that loving relationships are
typically sought for internal ends, either solely or in combination with
external ones. There is no natural end to partnership (or control, or other
internal ends): since they are internal to the relationship, they can only be
realized through entering and maintaining the relationship. I suspect that
the conditions of jus post amantes necessitudo will differ (but overlap)
depending on whether external goals have been achieved and/or internal
ones have been surrendered. This deserves attention that I cannot give
it here.

References
Ebels-Duggan, K. 2008. Against Beneficence: A Normative Account of Love.
Ethics 119: 142–170.
Fisher, M. 1990. Personal Love. London: Duckworth.
Frowe, H. 2011. The Ethics of War and Peace: An Introduction. London:
Routledge.
Ginet, C. 1990. On Action. Cambridge: Cambridge University Press.
Helm, B. 2017. Love. In The Stanford Encyclopedia of Philosophy (Fall 2017
Edition), ed. Edward N. Zalta. https://plato.stanford.edu/archives/fall2017/
entries/love/. Accessed February 14, 2020.
Lazar, S. 2020. War. In The Stanford Encyclopedia of Philosophy (Spring 2020
Edition), ed. Edward N. Zalta. https://plato.stanford.edu/archives/spr2020/
entries/war/. Accessed June 2, 2020.
Mele, A., ed. 1997. The Philosophy of Action. Oxford: Oxford University Press.
 6 NOT ALL’S FAIR IN LOVE AND WAR: TOWARD JUST LOVE THEORY 123

Moseley, A. 2020. Philosophy of Love. In The Internet Encyclopedia of Philosophy.

ISSN 2161-0002. https://www.iep.utm.edu/. Accessed January 28, 2020.
Scruton, R. 1986. Sexual Desire: A Moral Philosophy of the Erotic. New York:
Free Press.
Solomon, R.C. 1988. About Love: Reinventing Romance for Our Times. New York:
Simon & Schuster.
Stocker, M. 1976. The Schizophrenia of Modern Ethical Theories. The Journal of
Philosophy 73 (14): 453–466.
The Relapse, or Myrtle Bank. 1789. Multiple Authors. Farmington Hills, Michigan:
Gale Ecco, Print Edition 2018.
Velleman, J.D. 1999. Love as a Moral Emotion. Ethics 109: 338–374.
———. 2008. Beyond Price. Ethics 118: 191–212.
 CHAPTER 7

Doubting Love

Larry A. Herzberg

7.1   Introduction
In Graham Greene’s novel The End of the Affair,1 Maurice, the protago-
nist and narrator, recounts the following moment with Sarah, the married
woman with whom he’d been having a romantic love affair:

she said to me suddenly, without being questioned, ‘I’ve never loved any-
body or anything as I do you.’ …We most of us hesitate to make so com-
plete a statement—we remember and we foresee and we doubt. She had no
doubts. The moment only mattered. Eternity is said not to be an extension
of time but an absence of time, and it seemed to me that her abandonment
touched that strange mathematical point of endlessness… What did time
matter—all the past and the other men she may from time to time (there is
that word again) have known, or all the future in which she might be mak-
ing the same statement with the same sense of truth? When I replied that I
loved her too in that way, I was the liar, not she, for I never lose the

1
Thanks to Stewart Cole for suggesting this novel to me.

L. A. Herzberg (*)
University of Wisconsin, Oshkosh, Oshkosh, WI, USA
e-mail: herzberg@uwosh.edu

© The Author(s), under exclusive license to Springer Nature 125

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_7
126 L. A. HERZBERG

c­ onsciousness of time: to me the present is never here: it is always last year

or next week. (Greene 1951: 50–51)

Maurice’s insecurity and anxiety, which make it impossible for him to

relax and fully appreciate his liaisons with Sarah, are time-driven: they are
consequences of his inability to love in the present moment. As a result, he
can never fully accept the love Sarah expresses when she tells him that she
loves him, because (like “most of us”) he has doubts about anyone’s abil-
ity to sustain love. These doubts are based on his own experience and
beliefs about human nature, but in this particular case they are strength-
ened by his knowledge of Sarah’s history, for she has confided to him that
she’d had many prior affairs due to her husband’s inability to sexually
satisfy her. Her insistence to Maurice that her love for him is unprecedented
fails to convince him not because he thinks that she might be deceiving
him, but rather because he fears that she might be deceiving herself. After
all, despite her sincere professions of love to him, she had not yet given
him any indication that she was willing to leave her husband in the fore-
seeable future. Sarah, by contrast, is unconcerned about both the past and
the future. By loving entirely in the present, she can focus on love’s feel-
ings as they occur. This allows her to love not only without anxiety about
the future, but also without remorse, regret, or guilt—emotions usually
aimed at one’s past actions. Also, Sarah’s extraordinary way of loving is no
less rich for being focused exclusively on the current moment, for as
Maurice muses, if eternity is considered to be timelessness rather than ever-
lastingness, Sarah can love him eternally in the present moment.2 This,
Maurice suggests, allows her to love without the sort of doubt that relent-
lessly plagues him. But is that so? Even if Sarah can love entirely in the
present, does that really inoculate her against all doubts about the truth of
her own professions of love, or at least make such doubts avoidable?
Conversely, if Maurice can love only in the “ordinary” way, with one eye
on the past and the other to the future, does it really follow that, for him,
doubts about the truth of his own professions of love are unavoidable? My
answer to both of these questions will be “not necessarily”, and laying the
groundwork for that answer is the main task of this chapter. But before
further addressing the epistemological questions about belief and doubt

2
Greene’s point here seems to be inspired by Wittgenstein’s comment: “If we take eternity
to mean not infinite temporal duration but timelessness, then eternal life belongs to those
who live in the present” (1922: 6.4311).
 7 DOUBTING LOVE 127

that concern us, we must settle on a plausible conception of what love is.
Here I will describe what I take love to be only in general terms and fill in
relevant details as the chapter progresses.
My view of love has much in common with psychologist R. J. Sternberg’s
“triangular theory”.3 In both of our views love has three main compo-
nents, two of which (emotional intimacy and passion) are primarily felt or
affective, the third (“decision/commitment” for Sternberg) being primar-
ily cognitive and volitional.4 We agree that love’s “core” emotional feelings
include those of closeness, affection, care, and concern, but I further hold
that love’s emotional aspect includes dispositions to feel various other
sorts of emotion related to one’s beloved, given certain sorts of situa-
tions.5 For instance, to mention an example to which we will return later,
if I love you, I am probably disposed to feel anger on your behalf
whenever I judge that someone has unjustly insulted you, even if you
would not feel angry about it at all. Such “self-originating” emotional
dispositions contrast with any empathetic disposition I may also have to
share your emotions as I become aware of your interests and viewpoint.
Secondly, for the type of love that concerns us here, love’s passionate feel-
ings typically include those of sexual attraction toward one’s beloved, as
well as feelings of certain desires, such as the desire for companionship and

3
Sternberg is Professor of Human Development at Cornell University. See Sternberg
(1986, 1988) for the original formulations of his “triangular” theory of love; see Sternberg
(1997) for “construct validation” of the surveys he uses as measurement instruments. For a
non-technical introduction to his work, as well as a full listing of his scientific papers on love,
see https://lovemultiverse.com (accessed 21 October 2020). I am not the first philosopher
to have been impressed by his research; see de Sousa (2015: 80–84).
4
I further explicate these terms—particularly “volition”—below. Of the many philoso-
phers of love, Henry Frankfurt (e.g., 1999 and 2004) is perhaps best known for having
developed a “volitional” conception of love. However, by “love” Frankfurt means something
far more general than interpersonal love, and his use of “volitional” equivocates between two
distinct senses of the term. For an excellent critical discussion of Frankfurt’s views, see
Ferreira (2015). For an illuminating history of the idea of volition or willing, see
Davenport (2007).
5
Dispositions are tendencies defined in terms of manifestations, triggers, and masks.
Fragility is a commonly cited example. Breaking easily is the main manifestation of an item’s
fragility, but a fragile item might never break if events that would trigger the manifestation
(e.g., dropping) never occur, or if the manifestation is masked (e.g., if the item were wrapped
in bubble-wrap). Similarly, what is key to having a psychological disposition is that one would
psychologically react and so behave in a certain way were a triggering event to occur, absent
any masking conditions.
128 L. A. HERZBERG

the desires to love and to be loved.6 The satisfaction of these desires often
adds to the positive phenomenology of being in a loving relationship (for
instance, by grounding feelings of gratitude toward one’s beloved), but
when the passions are intense they can also generate negative emotions,
such as jealousy. Also, when such passions are stronger for one lover than
for the other, the imbalance can lead the “needier” lover to feel ashamed
and the “less needy” lover to feel resentful. Thirdly, love’s volitional
aspects include any conscious, voluntary decision one may make to behave
lovingly toward one’s beloved, as well as any disposition one may have to
so behave, regardless of whether it was established by one’s voluntary
decision or not. In other words, love’s volitional aspects include both
commitment-­making and being committed, where “being committed”
entails merely having a disposition to behave lovingly toward one’s
beloved, regardless of its cause.
That love involves feelings should be uncontroversial. Semantically
speaking, “love” is a perfectly acceptable answer to the question, “What
do you feel for me?”, and the “for me” here indicates that the relevant
feelings are not simple sensations like itches or burns, but rather are
directed at (or about) someone, and hence emotions. Emotions are felt
responses to mentally represented objects, events, persons, or situations
that are in some way significant to the emotional person.7 So love’s feel-
ings qualify as being emotional insofar as they are felt responses to one’s
representation of one’s beloved. Some passionate feelings, such as feelings
of sexual attraction elicited by representations of another’s body, can also
count as emotional.8 But if love consisted entirely of emotional and pas-
sionate feelings, it would be difficult to explain the defensiveness of a typi-
cal response to the question “Do you love me?” when it is posed in a
long-term romantic relationship: “Of course I love you. How could you even
6
Sternberg takes love’s passion component to include desires for “self-esteem, succorance,
nurturance, affiliation, dominance, submission, and self-actualization” (1986: 122), but
these seldom show up in his research, or in the research of others using his constructs. To the
extent that such desires contribute to love’s phenomenology, it is their felt satisfaction or
frustration that is relevant, not their mere existence. The same is true for the desire to love
and the desire to be loved, which cannot be considered constituents of love on pain of circu-
larity; rather, I view them as being common motivations to love and to enter into loving
relationships.
7
For more on this sort of view, see Ekman (1999), Lazarus (1991), Damasio (2004),
Prinz (2004), and Deonna and Teroni (2012). To understand how my view diverges a bit
from these, see Herzberg (2009, 2012, 2018).
8
See Herzberg (2019) for a defense of this claim.
 7 DOUBTING LOVE 129

ask me such a thing?” For such a response clearly indicates that the respon-
dent has interpreted the question as a sort of accusation, and accusing
someone is reasonable only if they can be held responsible for having acted
wrongly. The problem is that, generally speaking, we are not responsible
for our having or not having emotional or passionate feelings. Rather, we
are responsible for the voluntarily formed intentions that result from our
consciously deciding or willing to pursue some goal, and of course for any
actions that follow from these. Similarly, if love were merely affective it
would be difficult to explain the appropriateness of believing oneself to
have been betrayed by one’s beloved after they unexpectedly end the rela-
tionship, as Maurice believes himself to have been betrayed by Sarah when
she ends their affair without explanation.9 For one can betray (or renege
upon) only a commitment, agreement, or understanding that one has at
least implicitly made or entered into; one cannot in the same sense betray
a combination of feelings. These observations, along with others to be
discussed later, indicate that any credible view of love, and in particular of
romantic love, must include a volitional component.
But what, exactly, do I mean by “romantic”? Here I diverge a bit from
Sternberg’s use of the term, and to explain why it will be helpful to list the
eight “types of love” he generates from possible combinations of emo-
tional intimacy, passion, and decision/commitment. These are non-love
(no component present), liking (emotional intimacy only, the main ingre-
dient of friendship), infatuated love (passion only), empty love (decision/
commitment only), romantic love (emotional intimacy and passion), com-
panionate love (emotional intimacy and decision/commitment), fatuous
love (passion and decision/commitment), and consummate love (all com-
ponents present).10 With his inclusion of “non-love”, mere “liking”, and
“infatuated love” (his main example of which is having sex with a prosti-
tute), I think that it would be better to call this a non-exhaustive list of
relationship types rather than of love types. But more importantly, I think
that Sternberg is mistaken when he suggests that romantic love does not
include a significant level of commitment. He cites Shakespeare’s Romeo
and Juliet as portraying a paradigm case of romantic love, but here we
should recall one of Juliet’s most famous lines, which she addresses to
Romeo prior to their marriage: “O, swear not by the moon, the inconstant

9
Much of Greene’s novel consists of Maurice’s attempt to find an explanation for Sarah’s
apparently having betrayed him in this way. What he discovers is the novel’s major plot twist.
10
Sternberg (1986: 122).
130 L. A. HERZBERG

moon, that monthly changes in her circled orb, lest that thy love prove
likewise variable.”11 Surely this suggests that Juliet would take commit-
ment to be essential to the sort of love she wants to share with Romeo,
because it is needed to ensure constancy in a way that passion and emo-
tional intimacy by themselves cannot.12 Remarkably, even Sternberg notes
that consummate love, with its significant levels of all three components,
“is a kind of love toward which many of us strive, especially in romantic
relationships”.13 So I use the term “romantic love” to refer to what
Sternberg calls “consummate love”, and what he calls “romantic love”,
with its high levels of passion and emotional intimacy but negligible
amount of commitment, I call “sexual friendship”.
Obviously, there is much more to say about the components of roman-
tic love. In particular, I have not yet specified the contents of romantic
commitments, nor have I explained the important roles they play in
romantic relationships. Such details will be filled in as needed while we
address the epistemological questions that are our main concern. First, in
the section entitled “On Believing That I Love You: Two Potential Sources
of Bias”, I will outline two potential sources of bias that may cause one to
believe that one romantically loves another when one does not. Then,
partly on the basis of those potential sources of bias and partly on the basis
of more specific issues, in “On Believing That I Am Experiencing Love’s
Emotions Toward You” and “On Believing That I Am Making Love’s
Commitments to You”, I will argue that, at least to the extent that one is
aware of these issues, one may reasonably doubt that one is experiencing
romantic love’s emotional feelings (even when one is experiencing them),
and one may reasonably doubt that one is making love’s commitments
(even when one is making them). Finally, in “Concluding Remarks”, I
order by relative dubitability the propositions that must be true about
one’s passions, emotions, and commitments toward another in order for
one to romantically love them, and explain why doubts about these prop-
ositions are neither always avoidable when loving in Sarah’s extraordinary
way, nor necessarily unavoidable when loving in Maurice’s ordinary way.

11
Shakespeare (2014), Act II Scene II.
12
See Fehr (1988) for independent empirical support that the ordinary concepts of roman-
tic love and commitment significantly overlap. Fehr (559–560) takes her results to be incon-
sistent with Sternberg’s theory, but I think that this conclusion is based on the erroneous
assumption that in Sternberg’s view the concept of commitment is entirely contained in the
concept of love.
13
Sternberg (1986: 124), italics added.
 7 DOUBTING LOVE 131

7.2   On Believing That I Love You: Two Potential

Sources of Bias
Suppose you ask me whether I love you romantically, and I seriously con-
sider the question for the first time. I understand that whether I love you
in this way depends on whether my attitudes toward you include emo-
tional feelings of closeness, care, concern, and affection, passionate feel-
ings of sexual attraction, and various sorts of commitment. There are at
least two reasons for me to think that any belief I may now form that I love
you is unjustified. The first is that I might so desperately want to be loved
by someone I love that, given your apparent willingness to enter into such
a relationship with me, I may immediately develop a “confirmation bias”
that skews my judgments about my feelings and commitments toward
you.14 The second is that it is probably easier for me to identify my feelings
of sexual attraction toward you than it is for me to identify the various
types of my emotional feelings toward you,15 due to the more distinctive
and localized bodily conditions that feelings of sexual attraction register,
as well as the fact that there are more types of emotion to potentially con-
fuse, as will be illustrated in the next section. Of course, if I do first recog-
nize that I am sexually attracted to you, this gives me reason to believe that
I have met at least one of the conditions of romantically loving you. But it
also gives me another reason to think that I may be biased when making
judgments about how I otherwise feel toward you, as well as about my
level of commitment (or willingness to commit) to you. After all, if I know
that you are looking for romantic rather than infatuated or fatuous love,
the ongoing satisfaction of my sexual desire may well depend on my main-
taining a relationship with you that includes at least the expressions of emo-
tional intimacy and commitment. This may motivate me to behaviorally
simulate emotional intimacy and commitment, as some people simulate
sexual passion to ensure the satisfaction of their other needs. Importantly,
such simulation need not amount to intentional fakery. Particularly in the
case of emotional intimacy, I may fool myself as much as I fool you. For

14
The motivational power of a desire to be loved should not be underestimated. In
extreme cases, it can lead to horrendous behavior. This was noted by Patricia Krenwinkel, the
former “Manson Family” member: “It is countless how many lives were shattered by the
path of destruction that I was part of, and it all comes from just such a simple thing as just
wanting to be loved.” https://www.nytimes.com/2014/08/05/opinion/my-life-after-­
manson.html at 6:54, accessed 21 October 2020, italics added.
15
See Sternberg (1986: 120–123).
132 L. A. HERZBERG

passion-­simulators are more likely to be aware of their lack of passion than

emotion-­simulators are likely to be aware of their lack of emotion, given
that passion—or the lack thereof—is more accessible to conscious
awareness.16
So if I desire to be loved and I am sexually attracted to you, I initially
have at least two reasons to think that any judgment I make that I love you
may be biased and hence lack adequate justification. This bias could affect
both my introspective judgments concerning my present attitudes toward
you, and any inferences I might draw from my memories of how I’ve
behaved toward you.17 It is true that if I remember that I’ve expressed very
few of love’s feelings and commitments toward you, that may be evidence
that I do not love you. But my lack of expression could also be due to
quirks of my personality of which I am unaware; for instance, I might
simply tend to squelch any expressions of emotional intimacy due to inse-
curity. By the same token, if I remember that I have expressed many of
love’s feelings and commitments toward you, I have to allow that those
expressions may have been due to the sort of strongly motivated simula-
tion just discussed. Of course, I might also try to remember the relevant
feelings and voluntary acts of commitment-making themselves, rather
than merely their expressions. But using my memory here would seem to
be no more reliable than attempting to introspect my current mental states
and attitudes, despite additional concerns that may arise about the reli-
ability of that process. Since those additional concerns do not entirely
overlap in the cases of emotional feeling and commitment, I will explore
each of them in turn.

7.3   On Believing That I Am Experiencing Love’s

Emotions Toward You
Can I reliably type-identify, conceptualize, or “label” love’s emotions
based merely on the way they feel to me? For the moment, let’s assume
that I can always tell what my emotional feelings are about, so we can focus
only on the nature of the feeling itself. Let’s also assume that the qualita-
tive properties of such feelings can, like those of sense perceptions, be

16
Ibid.
17
Sternberg recognizes the possibility of bias in his subjects’ self-reports of their feelings
and commitments toward their partners due to their “tendencies to idealize their own rela-
tionships” (1997: 317).
 7 DOUBTING LOVE 133

embedded in phenomenal concepts and stored in long-term memory for

later use in recognizing the types of “incoming” emotional feelings.18 For
instance, suppose that as young children we have sensational experiences
that we can recognize based on their qualitative properties (how they feel to
us), and given the situations in which they occur, we learn from our lin-
guistic communities to label those types of sensations as feelings of fear,
anger, sadness, joy, and so on. Even on those assumptions, there are good
reasons to doubt the reliability of any process of emotion-type recognition
that rests exclusively on qualitative comparison. For ranges of sensations
can count as being of the same type, and the borders of those ranges tend
to be vague. This is clearest in the case of color sensations, where it may
be explained by the continuity of the light frequencies registered by the
retina, as well as by the visual system’s limitations when it comes to distin-
guishing between some number of adjacent frequencies. Think, for
instance, of the many shades of blue, and the narrower range where it
seems arbitrary to conceptualize a color sensation as being a shade of blue
or a shade of turquoise. Such indeterminacy and vagueness seem similarly
evident in the emotional case, where ranges of somatosensory sensations
can count as being of the same qualitative type, and we have good reason
to believe that there is even more vagueness between emotion types that
feel similar, given that the bodily conditions the somatosensory system
registers during emotion occurrences—heart rate, respiration, muscular
tension, hormone levels, and so on—substantially overlap between emo-
tion types. Even the most central instances of emotional feeling types
seem qualitatively similar to those of other types, including types relevant
to forming a justified belief that one loves another. For instance, in trying
to determine whether my feelings toward you are those of romantic love
or merely those of infatuated or fatuous love, it would be important for
me to figure out whether I tend to feel affection for you, or whether I tend
only to feel sexually attracted to you. But based merely on the ways they
feel, low levels of sexual attraction might be mistaken for moderate levels
of emotional affection; that is, a simmering level of passionate “heat”
might easily be misinterpreted as a moderate level of emotional “warmth”.19
18
See Chalmers (2003) for discussion of phenomenal concepts and Gertler (2001) for
discussion of how the “embedding” of qualia in concepts might take place.
19
Sternberg (1986: 122) distinguishes between emotional warmth, passionate heat, and
cognitive coldness. The temperature metaphor’s aptness is easy to explain if we view emo-
tional and passionate feelings as being somatosensory registrations of bodily conditions, and
cognition as being primarily neurological.
134 L. A. HERZBERG

But if I tend to experience only sexual attraction toward you, I do not love
you romantically.
Consider next a case in which recognizing what my feeling is about is
necessary for me to determine its type, but in which I am not able to dis-
criminate between two relevant alternatives. For instance, suppose that
shortly before you ask me whether I love you, you mention to me that
you’ve been feeling a bit ill, and this conversation is taking place during a
deadly pandemic. I immediately feel distressed, but I’m not sure whether
I am feeling concern about your condition or rather anxiety that I might
catch the disease from you. If I am experiencing the concern, I should prob-
ably count it as evidence that I love you (such concerns being among
love’s core emotions), but if I’m rather experiencing the anxiety, I proba-
bly should not; indeed, it might even provide me with evidence that I do
not love you. To the extent that concern and anxiety have similar qualita-
tive properties, it might be impossible for me to discriminate between the
two emotions based merely on how they feel. But how then can I tell that
my feeling is about you or about me? Especially if I have prior knowledge
that I tend to feel anxious about catching deadly diseases, I may well be
unable to justifiably infer that I am feeling concern about your condition,
even if I am.
There are also reasons to be concerned about the reliability of emotion-­
type identifications based on what one takes to have caused or elicited
one’s feeling, even when one has no trouble determining this. For instance,
suppose that what justifies inferences from an emotion’s cause to its type
is that emotion types are strongly associated with “paradigm scenarios” to
which they are normal or appropriate responses.20 More specifically, sup-
pose that any emotion felt in response to a situation that sufficiently
resembles a paradigm scenario of a given emotion type is highly likely to
be an emotion of that type. Using such a view by itself to justify emotion-­
type identifications fails to adequately allow for unusual emotional
responses due to atypical emotional dispositions. For instance, a paradigm
scenario of fright is suddenly being attacked by a dangerous predator.
However, “adrenaline junkies” often react to situations closely resembling
this with glee instead of fright; consider surfers thrilled rather than fright-
ened by the sudden approach of a thirty-foot wave. Furthermore, using
paradigm scenarios to justify emotion-type identifications does not ade-
quately take into account the fact that what caused an emotional feeling is

20
See de Sousa (1987) for a discussion of paradigm scenarios.
 7 DOUBTING LOVE 135

not always what the emotion is about. For instance, when you tenderly
express your love for me, I might know quite well that you are presenting
me with a paradigm scenario for my feeling affection toward you. However,
if my character is—perhaps unbeknownst to me—somewhat vicious, your
tender expression may elicit in me only an emotion of happiness that I can
now take advantage of you. In this case, what my happiness is about (that I
can now take advantage of you) is quite different from what a feeling of
affection would have been about (your tenderness). But, given the quali-
tative similarity of happiness and affection, if I happen to at least implicitly
accept the popular (but mistaken)21 view that what an emotion is about is
necessarily what the subject takes to have caused it, I may fail to even
notice what my emotion is actually about, and focus instead on the sce-
nario that I correctly take to have caused it. This may result in my errone-
ously believing that my happiness that I can now take advantage of you is an
emotion of affection for your tenderness. That is, in such a case I might get
wrong both my emotion’s type and that which it is about.
It might here be objected that an emotion’s type is never to be inferred
directly from its cause’s resemblance to a paradigm scenario, but rather
from how its cause has been evaluatively appraised by the subject. So if I
am in fact feeling happy that I can now take advantage of you, that emo-
tion must have been caused by my appraising your tenderness as an oppor-
tunity to exploit you, rather than as a gesture worthy of my affection.
However, this raises the question of the extent to which we are aware of
the appraisals that may cause our emotions. According to many emotion
theorists, such appraisals are seldom consciously and cognitively articu-
lated by the subject; rather, they usually occur automatically and uncon-
sciously. Indeed, “affect-program” theorists argue that the proper function
of an emotion is to prepare the subject to react to the causal event more
quickly than any consciously articulated cognitive appraisal would allow.22
In a similar vein, Jesse Prinz (2004) argues that in most cases the

21
The view that what an emotion is about is necessarily what the subject takes to have
caused it is fairly widespread among emotion theorists. For instance, Damasio (1994: 161)
seems to accept the view as a matter of psychological necessity. Prinz (2004: 62) appears to
accept it as a matter of semantic necessity. Lazarus (1991) and other causal-evaluative
appraisal theorists at least implicitly accept it when they hold that an emotion’s type is deter-
mined by the subject’s appraisal of its cause, and that the resulting evaluative judgment
remains a sustaining part of the emotion (providing it with direction). See Herzberg (2009)
for an extended argument that the view is mistaken.
22
Cf. Griffiths (1997) for a defense of affect program theories.
136 L. A. HERZBERG

a­ utomatically occurring emotional feeling just is the evaluative appraisal.

Such a view perhaps ensures that the subject is aware of the relevant
appraisal (since the appraisal just is the feeling), but it resurrects the prob-
lems already discussed about attempting to infer an emotion’s type from
its qualitative properties alone.
If my happiness case seems too far-fetched or pathological to be con-
vincing, the same points can be illustrated using a more normal case—one
that does not require me to have an atypical emotional character, or to
accept the view that one’s emotions are always about what one takes to be
their causes. Suppose that I hear someone say something negative about
you, and I am suddenly aware of feeling either anger or contempt (disdain)
toward the speaker—two emotions that feel quite similar to me. If it is
anger, I should probably count it as evidence that I love you, for, as I
mentioned in the introduction, feeling anger in such circumstances is
probably the manifestation of a “self-originating” disposition characteris-
tic of love. Anger can have this status because it is properly directed at
someone one appraises to have been unjustly offensive to oneself or to
someone to whom one feels close, including family members, close friends,
and lovers. Contempt, by contrast, is typically not felt on anyone’s behalf,
and is directed toward those one appraises to be unimportant or unworthy
due to their ineptitude, stupidity, or low social standing.23 Now, given that
it is the speaker’s remark to which I am reacting in either case, I must try
to determine whether I appraised the remark as unjustly offensive toward
you (in which case I should infer that I am feeling angry at the speaker), or
whether I appraised it merely as a sign of the speaker’s stupidity (in which
case I should infer that I am feeling contempt toward the speaker). If
immediately after hearing the remark but prior to experiencing the feeling
I happened to have consciously thought, “That’s unjustifiably insulting to
you,” I could perhaps strongly infer that my feeling is one of anger. But,
as noted earlier, emotions are probably rarely caused by such consciously
articulated appraisals. Of course, I could now deliberately re-appraise the
remark, but there is no guarantee that this would have the same effect as

23
These characterizations of anger and contempt are drawn from Aristotle’s uses of the
terms in Rhetoric (1954), Book II, chapter 2. The characterization of “anger” is also consis-
tent with Lazarus’ (1991) “core relational theme” of anger: “A demeaning offense against
me and mine”. Although some view contempt to be a blend of other emotions, I agree with
Ekman (1999) that it is more likely a “basic” emotion that evolved to help us navigate
through social hierarchies.
 7 DOUBTING LOVE 137

the automatic, unconscious process that originally caused the emotion.24

So, even if inferences from appraisal type to emotion type are reliable, it
can be difficult to determine the specific appraisal that actually caused
one’s emotion, especially when the emotional feeling’s qualitative proper-
ties fail to disambiguate between relevant alternative emotion types.25
Finally, we also need to recognize that many people simply misconcep-
tualize the types of their emotions because they have a less than perfect
grasp on the relevant emotion concepts. For instance, many people appear
not to understand the difference between jealousy and envy; in particular,
they tend to misconceptualize feelings of envy as being feelings of jealousy.
Such people are likely to say to someone who has bought something they
covet, “I’m so jealous of you!” They may have forgotten, or never learned,
the two emotions’ distinct analyses, jealousy being a response to a valued
relationship being threatened by a third party, envy being a response to
someone having something one wants. But if I tend to misconceptualize
feelings of envy as feelings of jealousy, I may falsely believe that I have an
emotional disposition characteristic of the sort of “ordinary love” Maurice
feels toward Sarah in The End of the Affair. For instance, suppose that you
have a close friendship with someone other than me that revolves around
your mutual interest in playing tennis, a game for which I have no apti-
tude. I might then come to believe that I am jealous of your friend (and
take my jealousy as a sign that I love you), when in fact I am merely envi-
ous of them for having an ability that enables the two of you to engage in
a fun activity closed to the two of us.
These are just a few of the ways in which one can be wrong about the
types of one’s emotional feelings, particularly in regard to the feelings of
love. I am not suggesting that in general the underlying processes are so
unreliable that everyone always has sufficient reason to doubt that they
experience love’s emotions toward another. Rather, my point is merely
that some people under some psychological circumstances seem likely to
form false beliefs that they emotionally love another, and so have sufficient
reason to doubt that they do so (at least insofar as they are aware of the
relevant circumstances). Let’s assume for the sake of further exploration
24
Multilevel appraisal theories contrast conscious, deliberate, cognitive appraisals with the
automatic, unconscious appraisals made by the sub-personal emotional system. Doing so
helps to explain phobic emotions, among other phenomena. Cf. Teasdale (1999) for a con-
cise overview of such theories.
25
See Herzberg (2016) for my positive view of how we may be able to reliably identify the
types of our emotions.
138 L. A. HERZBERG

that I am one of those people. Upon reflection, I’m unsure whether I love
you or not, emotionally speaking; my confidence level that I tend to expe-
rience love’s emotions toward you is below 50%. Should I therefore doubt
that I love you romantically? Perhaps, but let’s further suppose that I am
no more confident that I usually do not experience love’s emotions toward
you. In other words, I have insufficient evidence to justify either belief or
doubt that I love you. This allows me to suspend both belief and doubt
and to merely “entertain the hypothesis” that I love you, pending further
evidence. However, to gain more evidence I need to remain close to you,
so I consider whether to now make love’s commitments to you and to
express my doing so by telling you that I love you. After all, in contrast to
my feelings, whether I make commitments or not is entirely under my
control, right? Admittedly, the love that results might be empty (commit-
ment only) or at best fatuous (commitment and passion), but some
“arranged marriages” provide evidence that what begins as empty or fatu-
ous love can become romantic over time.26 So let’s suppose that I now
exclaim “I love you!” sincerely believing that I am making love’s commit-
ments to you. Does it follow that I am making those commitments to you,
and hence that I love you in at least that limited way?

7.4   On Believing That I Am Making Love’s

Commitments to You
Unfortunately, not quite. But before I explain why, I need to be clearer
about what I take to be the main commitments of romantic love—those
that are implied by one’s sincerely stating “I love you” in a romantic con-
text. Since in many cultures romantic love provides a good ground for
marriage (“Because we love each other” being a perfectly acceptable
answer to “Why are you getting married?”), and in many contemporary
cultures such love might even be considered necessary for a successful
marriage, marriage vows can provide some guidance here. Such vows
often require the couple to pledge monogamous fidelity to each other,
regardless of future circumstances.27 They also stress the intended irrevo-
cability of marriage commitments at least until the death of one of the

26
See Sternberg (1986: 123). Remember that I am using “romantic” as Sternberg uses
“consummate”, the type of love that requires adequate degrees of all three components.
27
For marriage vows from many religious traditions, see https://www.theknot.com/con-
tent/traditional-wedding-vows-from-various-religions (accessed 10/22/2020).
 7 DOUBTING LOVE 139

spouses. Some also involve pledges to demonstrate such virtues as honesty,

respect, and forgiveness toward one’s partner, which is reasonable insofar
as the exercise of such virtues involves voluntary activity; one is not implau-
sibly pledging to experience feelings over which one has little if any con-
trol. By contrast, non- or premarital romantic lovers usually are not
committing to a lifelong relationship when they tell each other “I love
you”, nor are they committing to share their lives with each other to the
extent found in marriage; for example, they need not commit to cohabita-
tion. However, telling someone that you romantically love them does
imply that you intend the relationship to last for at least some time; in
contrast to mere infatuation, romantically loving someone seems incom-
patible with intending merely to have a “one night stand” with them.
Romantic lovers also implicitly commit themselves to being accessible to
each other on an ongoing basis, in a way that resembles the “to have and
to hold” clause of some marriage vows. More specifically, they commit
themselves to being open to ongoing emotional intimacy and sexual activ-
ity (the other two components of romantic love), to a degree that exceeds
mere sexual friendship, but which may fall short of what is expected in an
ideal marriage. They also commit to a significant degree of practical
dependability, to come to each other’s aid and to prioritize each other’s
interests over those of their mere friends. They perhaps commit to being
honest about any firm intentions they may form to discontinue the rela-
tionship. Finally, while two romantic lovers need not commit to romanti-
cally loving only each other, de facto “fidelity” typically results from the
practical difficulty of maintaining with multiple partners the levels of emo-
tional and sexual accessibility just mentioned.28
Importantly, the commitments one makes to one’s romantic partner,
like those to one’s spouse, are addressed to their singular individuality;
they are not conditional on one’s beloved’s maintaining inessential prop-
erties. As Shakespeare famously put it, “Love is not love which alters when
it alteration finds.”29 By contrast, while one’s friends are certainly valued,
they usually are valued for their properties. As a result, friends are inter-
changeable in a way that lovers are not. For instance, if one of my friends
can no longer meet me for a hike, another friend who is equally competent

28
I do not rule out the possibility of polyamorous romance, but for simplicity’s sake I am
concerned here with bilateral relations.
29
Shakespeare (2004), Sonnet 116. For some disagreement on this point, see Rorty
(1986/1993).
140 L. A. HERZBERG

at hiking (and perhaps at conversing) will do just as well. Similarly, if one

is merely sexually attracted to someone, one should be equally attracted to
their identical twin as well, ceteris paribus. Indeed, if one were not so
attracted to each twin, one might reasonably be viewed as inexplicably
fickle. But there is nothing similarly incoherent about romantically loving
one twin but not the other, and this can be explained by the fact that love’s
commitments are addressed to the singular individual who is loved.30 This
also helps to explain the exceptional degree of value most people place on
being loved, as opposed to merely being admired for their properties or
their accomplishments. For they understand that through another’s love,
their singular individuality is recognized and affirmed in a way not found
in other forms of relationship. Also, the exceptional degree of value one
places on being loved in this way explains why many spouses and romantic
lovers would prefer their partners to have casual sex with a stranger rather
than with a sexual friend with whom they may share an emotional inti-
macy, and with a sexual friend over a competing lover to whom they may
be committed. For those other forms of relationship are less threatening
to the commitments that hold between spouses or romantic lovers qua
singular individuals. Finally, none of this entails that one’s emotional and
passionate responses to another’s properties play no role in helping one to
select a lover. The fact that your particular properties trigger in me certain
positive feelings may well be why I decide to address love’s commitments
to your singular individuality rather than to someone else’s, even though
everyone is a singular individual. But although one’s emotional and sexual
responses to another’s properties help to explain romantic love’s selectiv-
ity, an important function of commitment is precisely to transcend the
relative shallowness of such selectivity.
Mutual commitment plays another important role in romantic relation-
ships: it allows the lovers to form a union that amounts to a merging or
interlinking of their interests. For if I value your commitments to me
(which I should, given my desire to be loved and willingness to be loved
by you), I thereby have an interest in your maintaining your commitment
to me. My recognition of this interest should motivate me to make and
maintain a reciprocal commitment to you that you similarly value and rec-
ognize, providing you with an interest in my maintaining my commitment
to you. I should then also value the maintenance of my commitment to

30
This is what philosophers mean when they observe that love is not “fungible”. See, for
instance, de Sousa (1987).
 7 DOUBTING LOVE 141

you at least insofar as it acts as a means of maintaining your commitment to

me, and vice versa.31 Of course, each of us may also have an independent
interest in our commitments to each other, insofar as we each desire to love
in addition to desiring to be loved. But commitments motivated only by a
desire to love would not require any relationship, and hence would be
unlikely to generate one. Finally, mere mutuality of affect (be it passionate
or emotional) seems unlikely to generate a stable relationship or merging
of interests, since one’s passions can be satisfied—and one’s emotions can
be elicited—by anyone with the relevant properties. So it seems that only
mutual commitments addressed to each other’s singular individuality and
motivated by each partner’s desire to be loved can create the sort of
interest-­merging that results in the maintenance of a loving relationship.
Note that once this interlinkage of interests is in place, each lover also has
a stake in helping their beloved pursue whatever interests they may have
external to the relationship, insofar as their doing so should strengthen the
other’s commitment to them, and as a result their commitment to
the other.32
Let’s now return to our epistemological question: does it follow from
my merely believing that I am making love’s commitments to you that I
am making them? Is this an aspect of loving about which doubt is always
unreasonable? As I stated earlier, not quite. For even if any commitment I
make is entirely under my control, it does not follow that any belief I have
that I am now making love’s commitments to you must be true. What may
make it seem otherwise is the conceptual truth that one voluntarily makes
a commitment if and only if one knows that one is doing so. Call this “C”. It
follows trivially from the truth condition of knowledge that if one knows
that one is making a commitment, then one is doing so. And it follows
from the definition of “voluntarily” that if one voluntarily makes a com-
mitment, then one knows that one is doing so. For one cannot voluntarily
do anything without knowing what one is doing. To put this somewhat

31
This is consistent with Robert Solomon’s observation that the “grand reason” to love “is
because we bring out the best in each other” (1988: 155). My love for you brings out the
best in me insofar as I want to be the best person I can be in your eyes in order to reinforce
your commitment to me, and vice versa.
32
Notice that on this account of commitment and interest-merging in a romantic relation-
ship, the merger does not generate a new “we-entity” separate from each of the lovers’ sin-
gular individualities. See Helm (2017) for discussion of such views. Rather, on my analysis,
it is essential that each lover maintain their autonomous ability to withdraw their commit-
ment to the other.
142 L. A. HERZBERG

differently, any voluntary act implicates the agent; this is why one can
always be held responsible for one’s voluntary actions, including one’s
commitment-makings. Finally, since one cannot make a commitment
without knowing that one is doing so, and one cannot know that one is
doing so without justifiably believing that one is doing so (since knowing
entails justifiably believing), the beliefs that are partly constitutive of acts of
commitment-making are both justified and made true by those acts them-
selves. However, little of epistemological importance follows from C. For
like all conceptual truths, C fails to settle any non-conceptual factual
issues. In particular, it does not follow from C that just any belief I may
have that I am making a commitment is true, for such a belief might be
produced in any number of ways other than by my commitment-making.
For instance, a sufficiently crafty neurologist might implant such a belief in
me despite it being false. Or, more realistically, I may hold the belief as a
result of hypnosis, or I might form such a belief due to wishful thinking or
the two potential sources of bias discussed earlier. In other words, C does
not provide me with any means for discriminating true beliefs that I am
making a commitment from false beliefs that I am doing so.
Someone might here object that there is no need to discriminate true
from false beliefs that one is making a commitment, because to believe that
one is making a commitment just is to make the commitment. Indeed, the
objector might claim, merely uttering the words “I hereby make this com-
mitment” makes it so. The objector’s strategy here is to assimilate
commitment-­making to a merely performative analysis of promising. On
such an analysis, if I say to you, “I promise to have dinner with you tomor-
row,” my saying the words makes it true that I have promised to have din-
ner with you tomorrow. It does not matter whether I am stating the words
sincerely or not. Indeed, even if I am being thoroughly deceptive when I
say “I promise you…”, I nevertheless have made the promise to you. If
commitment-making is similarly performative, then my merely saying “I
am making love’s commitments to you” is criterial of my having made
those commitments. So if I merely say to you “I love you”, understanding
the commitments that love requires, I thereby in fact love you (at least in
Sternberg’s empty sense). But surely that is not the case. Rather, the truth
of my words depends on whether I am willfully making love’s commit-
ments to you—that is, on whether I am forming an intention to keep those
commitments (even if that intention goes unfulfilled). So if the objector
were to insist that promising and commitment-making must share a single
analysis, I would argue that we should give up the performative analysis of
 7 DOUBTING LOVE 143

promising in favor of a volitional one, rather than give up the volitional

analysis of commitment-making in favor of a performative one.33
However, the epistemological problem of how I can tell my true from
my false beliefs that I am making a commitment may not be as serious as
it appears, for two reasons. First, it seems likely that there is a phenomeno-
logical difference between my believing that I am making a commitment
when my will is engaged versus my so believing when my will is not
engaged. That is, there seems to be a feeling of resolve that accompanies the
making of a commitment, and perhaps I can reliably tell when that feeling
is present or absent. Note that resting the ability to tell true from false
beliefs that I am making a commitment on my being able to correctly
identify a feeling of resolve need not open up a can of worms similar to the
ones that supplied reasons to doubt that I was feeling love’s emotions. For
in the emotional cases, the problems all involved the difficulty of discrimi-
nating between different types of emotion that felt qualitatively similar,
while in the commitment case there seems not to be different types of will-
ing that could feel qualitatively similar. Secondly, and more importantly,
even if I do not know how I can reliably distinguish true from false beliefs
that I am making a commitment, it does not follow that I cannot in fact
do so, and so my mere lack of knowledge in this regard does not provide
me with a positive reason to doubt that I can distinguish them.
This second point can be further elaborated by noting an epistemologi-
cal principle that applies to both beliefs about one’s commitment-makings
and beliefs about the types of one’s emotions: absent good reason for judg-
ing such propositions to be false, one is justified in accepting them as true “by
default”. In this respect, introspectively and reflectively produced beliefs
resemble perceptually produced beliefs about the presence and types of
physical objects in the world.34 Both are similarly “foundational”, episte-
mologically speaking.35 If I perceptually believe that there is a cat on the
mat, I am justified in believing that proposition unless my default

33
The paradigm of a performative practice is the “christening” or naming of a ship:
exclaiming, for instance, “I hereby christen this ship the Santa Maria!” Such a paradigm
seems quite distant from the practice of promising.
34
Note that this claim does not require that we view mental states as otherwise analogous
to physical objects, nor does it require introspectively and perceptually formed beliefs to have
default justifications of the same type or strength. This foundational sort of justification can,
of course, be supplemented by a belief’s consistency and coherence—and defeated by its
inconsistency and incoherence—with other beliefs.
35
See, for instance, Audi (2002) for his version of fallibilistic foundationalism.
144 L. A. HERZBERG

justification for it is defeated by some good reason to think it is false, such

as that my perceptual or cognitive abilities are malfunctioning due to unfa-
vorable conditions of some sort. Similarly, if I introspectively believe that
I am making love’s commitments to you, I am justified in believing that
proposition unless my default justification for it is defeated by some good
reason to think it is false, such as that my introspective and cognitive abili-
ties are malfunctioning or are biased by the sorts of factors mentioned in
“On Believing That I Love You: Two Potential Sources of Bias”. The
same point applies to my introspectively produced beliefs that I am expe-
riencing a particular type of emotion toward you, only in this case there
potentially are additional reasons for doubt of the sort outlined in “On
Believing That I Am Experiencing Love’s Emotions Toward You”. It is
important to stress that only beliefs produced primarily by foundational
processes like introspection and perception can enjoy this sort of default
justification, and that such justification certainly does not guarantee that
the believed propositions are true. However, it does rule out one’s being
justified in doubting the same propositions at the same time (absent some
evidence of their falsity).
Of course, even when one is justified in believing that one is making
love’s commitments, one cannot be similarly justified in believing that one
will remain committed between those acts of commitment-making, nor
can one be similarly justified in believing that one will keep the commit-
ments one has made. For such predictive beliefs can be justified only by
inference from what one justifiably believes about one’s own history, and
in this way they are similar to the beliefs one might have about one’s
beloved’s ability to keep their commitments. In many cases, the more one
knows about the relevant history, the more reasonable it may be to doubt
that a commitment will be kept. Ideally, making a commitment establishes
a disposition to behave consistently with its content until the commitment
has been revoked by the agent. But given the less-than-perfect reliability
with which such dispositions operate, as well as the fact that psychological
dispositions are not directly observable and can be inferred only through
observations of their manifestations, the best evidence I may have that I
remain committed might be behaviors that others could observe and
assess at least as reliably as I. So as long as I keep in mind that my overt
expressions of love could be misleading, others’ estimations of my charac-
ter might provide me with a valuable “reality check” on the nagging ques-
tion of whether my love is true.
 7 DOUBTING LOVE 145

7.5   Concluding Remarks

Although we have merely scratched the surface of the conceptual and psy-
chological aspects of romantic love and the epistemological issues related
to justifiably believing or doubting that one loves another, it seems safe at
this point to draw a few conclusions. To begin with, one can certainly be
wrong about whether one romantically loves another. That is, one’s beliefs
on this matter are clearly fallible. However, fallibility does not by itself
entail dubitability. One is justified in believing that one loves another
unless one has good reason to doubt the reliability of one’s introspective,
reflective, or inferential processes, and the potential reasons for doubt can
vary from person to person, case to case, and target only propositions
about one having particular components of love, some of which seem
more readily dubitable than others.
We can now order by their relative dubitability the propositions that
must be true about one’s passions, emotions, and commitments toward
another in order for one to romantically love them. First, one may be best
situated to justifiably judge whether one is experiencing passionate feel-
ings of sexual attraction to another, thanks to their distinctive qualitative
profiles, which can be explained by the particular bodily conditions they
register. There may also be other subjectively observable mental signs of
sexual arousal and attraction, such as distinctive forms of attention, per-
ceptual focus, and imagery, which we cannot delve into here. So whatever
doubts one may have that one is experiencing feelings of sexual attraction
toward another are probably rarely justified. Secondly, one appears to be
fairly well-situated to justifiably judge whether one is making love’s com-
mitments to another, at least if there is a distinctive feeling of resolve pro-
duced by the engagement of one’s will that allows one to tell when one is
really making a commitment. If such feelings exist, identifying them
should at least not fall prey to the sorts of problems that can diminish
one’s ability to recognize one’s emotional feelings. After all, there is pre-
sumably only one type of feeling of resolve that can be produced by will-
ing, unlike the many types of qualitatively similar emotional feelings that
can be produced by a wide variety of situations with a number of poten-
tially emotion-eliciting properties. On the other hand, unlike emotional
feelings, feelings of resolve seem not to be associated with the sorts of
publicly observable “paradigm scenarios” that may facilitate one’s ability
to conceptualize or linguistically label one’s emotional feelings (despite
the problems that can arise when inferring emotion type from scenario
146 L. A. HERZBERG

type). So the epistemological usefulness of feelings of resolve remains

somewhat indeterminate, and here we may have to rely on the general
principle that introspectively produced beliefs are justified absent good
reasons for doubting their contents, such as reasons to believe that the
potentially biasing factors outlined in “On Believing That I Love You:
Two Potential Sources of Bias” are operative. Finally, it seems clear that
one is least well-situated to justifiably judge whether one is experiencing
love’s emotional feelings toward another, given the many ways one can
misconceptualize their types outlined in “On Believing That I Am
Experiencing Love’s Emotions Toward You”, in addition to the two
potential sources of bias. But here again, one’s introspectively produced
beliefs about the types of one’s emotional feelings are justified by default,
absent good reasons to think that their contents are false.
I can now explain why Maurice is not necessarily right to suggest that
doubt is always avoidable when loving in Sarah’s extraordinary way and
always unavoidable when loving in his own ordinary way. For even if Sarah
loves only in the present, she could still have justified doubts about the
types of her emotions and even about the truth of her apparent
commitment-­making. She might be insensitive to feelings of resolve, or
her will may not produce them with sufficient intensity, or she may be
biased by her sexual attraction to Maurice or by her intense desire to be
romantically loved. Indeed, the only sort of doubt that Sarah may always
be able to avoid is about commitment-keeping, insofar as she is uncon-
cerned with the future. Of course, one may wonder whether she can make
any commitment at all, given that she loves entirely in the present, and
commitments are essentially future-directed. But Sarah’s way of loving is
not incoherent; it merely represents a compromise with human imperfec-
tion. As we have noted, no commitment’s making can guarantee its being
kept; any commitment can be revoked for good reason, irrational influ-
ences on one’s will can result in inconsistent willings, and psychological
dispositions to behave in various ways are not fail-safe. Sarah, like anyone
else, can make a commitment by forming the necessary intention to keep
it in the future, without being concerned about keeping it in the future.
Maurice, by contrast, can experience only the sort of ordinary love that
breeds resentment and hatred when its commitments are unexpectedly
revoked, and insecurity or jealousy even when they are kept. Perhaps the
best he can do when Sarah proclaims her extraordinary love for him is not
to deny the value of her commitment-making at the moment it occurs,
however justified his concerns may be about her ability to keep it. Of
 7 DOUBTING LOVE 147

course, when he loves in his ordinary way, he will probably be more prone
to doubts about all of love’s components. His anxiety about the future
may undermine his ability to experience love’s emotional and passionate
feelings in the present, even when he may have been able to feel them
otherwise. Similarly, the doubts generated by his perhaps justifiable anxi-
ety might themselves interfere with his ability to experience love, and so
become self-fulfilling. However, not everyone who can love only in
Maurice’s ordinary way must meet such an unhappy fate. If one is lucky,
one’s history might provide no strong grounds for doubt, one might be
able to counterbalance anxiety with hope, and one might have no good
reason to ever doubt that one is experiencing love’s feelings. In the end,
despite the causal relations that can occur between them, what may be
most important to recognize is the fundamental independence of our pas-
sions, emotions, and commitments from whatever beliefs or doubts we
may form about them. It is one thing to love, and quite another to believe
or doubt that one loves. As long as we keep that firmly in mind, our
doubts are at least less likely to interfere with our loves.

References
Aristotle. 1954. Rhetoric. Trans. Rhys Roberts. New York: Modern Library.
Audi, R. 2002. The Foundationalism-Coherentism Controversy: Hardened
Stereotypes and Overlapping Theories. In Knowledge & Inquiry, ed. W.B. Wray,
83–127. Peterborough: Broadview Press.
Chalmers, D. 2003. The Content and Epistemology of Phenomenal Belief. In
Consciousness: New Philosophical Perspectives, ed. Q. Smith and A. Jokic,
220–272. Oxford: Clarendon Press.
Damasio, A. 1994. Descartes’ Error: Emotion, Reason and the Human Brain.
New York: Avon Books.
Damasio, A. 2004. William James and The Modern Neurobiology of Emotion. In
Emotion, Evolution, and Rationality, eds. D. Evans and P. Cruse, 3–13. New
York: Oxford University Press (USA).
Davenport, J. 2007. Will as Commitment and Resolve. New York: Fordham
University Press.
De Sousa, R. 1987. The Rationality of Emotion. Cambridge, MA: MIT/Bradford.
———. 2015. Love – A Very Short Introduction. New York: Oxford University Press.
Deonna, J., Teroni, F. 2012. The Emotions – A Philosophical Introduction, London:
Routledge.
Ekman, P. 1999. Basic Emotions. In Handbook of Cognition and Emotion, ed.
T. Dalgleish and M. Power, 45–60. Sussex, UK: John Wiley & Sons, Ltd.
148 L. A. HERZBERG

Fehr, Beverley. 1988. Prototype Analysis of the Concepts of Love and Commitment.
Journal of Personality and Social Psychology 55 (4): 557–579.
Ferreira, J.M. 2015. Willing and the Necessities of Love in Frankfurt and
Kierkegaard. In Love, Reason, and Will: Kierkegaard After Frankfurt, ed.
A. Rudd and J. Davenport, 197–216. New York: Bloomsbury Academic.
Frankfurt, H. 1999. Autonomy, Necessity, and Love. In Necessity, Volition, and
Love, 129–141. Cambridge: Cambridge University Press.
———. 2004. The Reasons of Love. Princeton, NJ: Princeton University Press.
Gertler, B. 2001. Introspecting Phenomenal States. Philosophy and Phenomenological
Research LXIII (2): 305–328.
Greene, G. 1951. The End of the Affair. New York: Penguin Books.
Griffiths, P. 1997. What Emotions Really Are: The Problem of Psychological
Categories. Chicago: University of Chicago Press.
Helm, B. 2017. Love. In The Stanford Encyclopedia of Philosophy, ed. Edward
N. Zalta. https://plato.stanford.edu/archives/fall2017/entries/love.
Accessed 22 October 2020.
Herzberg, L. A. 2009. Direction, Causation, and Appraisal Theories of Emotion.
Philosophical Psychology 22 (2): 167–186.
———. 2012. To Blend or to Compose: A Debate About Emotion Structure. In
Dynamicity in Emotion Concepts, ed. P.A. Wilson, 73–94. Lodz: Studies in
Language 27, Frankfurt a. Main: Peter Lang.
———. 2016. On Knowing How I Feel About That: A Process-Reliabilist Analysis.
Acta Analytica 31 (4): 419–438.
———. 2018. Can Emotional Feelings Represent Significant Relations?. Acta
Analytica, 34 (2): 215–234.
———. 2019. On Sexual Lust as an Emotion. Humana.Mente Journal of
Philosophical Studies 35: 271–302.
Lazarus, R.S. 1991. Emotion and Adaptation. New York: Oxford University Press.
Prinz, J. 2004. Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford
University Press.
Rorty, A.O. 1986/1993. The Historicity of Psychological Attitudes: Love Is Not
Love Which Alters Not When It Alteration Finds. In Friendship: A Philosophical
Reader, ed. N.K. Badhwar, 91–107. Ithaca, NY: Cornell University Press.
Shakespeare, W. 2004. Shakespeare’s Sonnets. Eds. B.A. Mowat and P. Werstine.
New York: The Folger Shakespeare Library.
———. 2014. Romeo and Juliet. Revised 4th ed. Cambridge: Cambridge
University Press.
Solomon, R. 1988. About Love: Reinventing Romance for Our Times. New York:
Simon & Schuster.
Sternberg, R.J. 1986. A Triangular Theory of Love. Psychological Review 93
(2): 119–135.
 7 DOUBTING LOVE 149

———. 1988. Triangulating Love. In The Psychology of Love, ed. R.J. Sternberg
and Michael L. Barnes, 119–138. New Haven: Yale University Press.
———. 1997. Construct Validation of a Triangular Love Scale. European Journal
of Social Psychology 27: 313–335.
Teasdale, J.D. 1999. Multi-Level Theories of Cognition-Emotion Relations. In
Handbook of Cognition and Emotion, ed. T. Dalgleish and M. Power, 665–681.
Sussex, UK: John Wiley & Sons, Ltd.
Wittgenstein, L. 1922/1961. Tractatus Logico-Philosophicus. New York:
Routledge Classics.
 CHAPTER 8

Love and Free Agency

Ishtiyaque Haji

8.1   Introduction
In this chapter, I motivate and explain the significance of the view that
love is historical in a sense of “historical” to be explained. I argue that
love’s historicity exposes its freedom presuppositions. These, in turn, ren-
der love fragile insofar as lack of free agency compromises love, lovable
behavior, or relationships of love.

8.2   Responsibility and History

To get a handle on the relevant concept of historicity, it will be instructive
to start with a prominent consideration in favor of moral responsibility
being historical. On a customary understanding of moral responsibility, an
agent being morally responsible for performing an intentional action
entails that she deserves some moral credit (e.g., praise) or discredit (e.g.,
blame) for that action. Various conditions of moral responsibility—agency
and control (or freedom) conditions, for instance—cannot be specified
independently of invoking elements of a person’s psychology, such as apt

I. Haji (*)
University of Calgary, Calgary, AB, Canada
e-mail: ihaji@ucalgary.ca

© The Author(s), under exclusive license to Springer Nature 151

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_8
152 I. HAJI

beliefs and desires. Call such elements “responsibility-grounding mental

elements” and dub conditions of responsibility that essentially appeal to
these elements “responsibility’s psychology implicating conditions.”
Responsibility historicism (or externalism) is the thesis that the psychology
implicating conditions of moral responsibility cannot be specified inde-
pendently of facts about how the person acquired her responsibility-­
grounding mental elements. The salient idea is that facts about one’s
history or past that bear on the acquisition of one’s responsibility-­
grounding mental elements can influence whether one’s actions are free
and, hence, are pertinent to whether one can be morally responsible for
them. Responsibility anti-historicism (or internalism) is the denial of
responsibility externalism.
In the free will literature, various vignettes involving surreptitious psy-
chological manipulation provide strong impetus for responsibility histori-
cism. In a case Alfred Mele advances, Chuck is morally responsible for
making himself a merciless bully of vulnerable people. He stalks and kills a
homeless man, Don (Mele 2019: 19–20).1 Sally, in contrast, works hard at
and succeeds in becoming one of the kindest, gentlest people on Earth,
and she is morally responsible for doing so. In case One Bad Day,

Sally awakes with a desire to stalk and kill a neighbor, George… What hap-
pened is that, while Sally slept, a team of psychologists that had discovered
the system of values that make Chuck tick implanted those values in Sally
after erasing her competing values. They did this while leaving her memory
intact, which helps account for her surprise… Seeing nothing that she
regards as a good reason to refrain from stalking and killing George, pro-
vided that she can get away with it, Sally devises a plan for killing him; and
she executes it—and him—that afternoon… When Sally falls asleep at the
end of her horrible day, the manipulators undo everything they had done to
her. When she awakes the next day, she is just as sweet as ever and she has no
memory of the murder. (Mele 2019: 20–21)2

Mele judges that whereas Chuck is morally responsible for his killings,
he “cannot help but see Sally as too much a victim of external forces to be
morally responsible for killing George” (Mele 2019: 21–22).3 I agree.4

1
See also, Mele 2006: 171, 2016: 73.
2
See also Mele 2006: 171–172, 2016: 73–74.
3
See also Mele 2006: 172, 2016: 75–76.
4
See, for example, Haji 1998: 15–22, 2009: 61–67, 2013; Haji and Cuypers 2008: 15–41.
 8 LOVE AND FREE AGENCY 153

This judgment is plausible even if it were not human manipulators but

some non-agential external force, a weird substance in the atmosphere, or
an interplay of Sally’s genetic profile and environmental factors that
brought about the radical change in Sally. If you concur with this judg-
ment, you should find responsibility externalism plausible. You should be
pretty convinced that how you acquire your responsibility-grounding
mental elements can make a difference to whether you’re morally praise-
worthy or blameworthy for your pertinent choices or bodily actions.

8.3   Love and History

Is love historical? It would seem that whether it is depends partly on the
nature of love. Central to some of the competing accounts of love are
desires. For example, Harry Frankfurt submits that lovers identify the
interests of their beloveds as their own; love is concerned with the well-­
being and flourishing of the beloved object. The good of the beloved is
desired for its own sake rather than for the sake of promoting other inter-
ests (Frankfurt 1999: 165–166). Stressing motivation, Frankfurt also
maintains that love is “essentially a somewhat non-voluntary and complex
volitional structure that bears both upon how a person is disposed to act
and upon how he is disposed to manage the motivations and interests by
which he is moved” (Frankfurt 1999: 165). Derk Pereboom affirms that
“love of another involves, most fundamentally, wishing well for the other,
taking on many of the aims and desires of the other as one’s own, and a
desire to be together with the other” (Pereboom 2001: 202). Harvey
Green submits that love involves a desire to share an association with the
beloved, and that the “basic desire for association motivates and sets
parameters for the desire for the good of the one who is loved” (Green
1997: 217).5
In other accounts of love, emotions figure centrally. Niko Kolodny ten-
ders that love essentially involves “emotional vulnerability.” He explains:

To say that A is emotionally vulnerable to B…is to say, in part, that A is

disposed to have a range of favorable emotions in response to A’s beliefs that

5
Indeed, Green’s position is stronger. Regarding romantic love, he claims that A loves B
if, and only if, A desires to share an association with B which typically includes a sexual
dimension, A desires that B fare well for his or her own sake, and A desires that B reciprocate
the desires for association and welfare (Green 1997: 216).
154 I. HAJI

B…has fared or will fare well, and a range of unfavorable emotions in

response to A’s beliefs that B…has fared or will fare poorly. For example, A
may feel content when B is well, elated when B meets with unexpected good
luck, anxious when it seems that B may come to harm, grief-stricken when
B does. (Notice that A is not simply emotionally vulnerable to how B treats
A, although this is often what is meant by saying that one person is “emo-
tionally vulnerable” to another.) (Kolodny 2003: 152)

In his complex Kantian analysis of love, David Velleman emphasizes

that love disarms our emotional defenses, making us vulnerable to others.
When we love someone, “we are responding to the value that he possesses
by virtue of being a person or, as Kant would say, an instance of rational
nature” (Velleman 1999: 365). Following Kant, Velleman claims that the
value of a person is different in kind from the value of other things: “a
person has a dignity, whereas other things have a price” (Velleman 1999:
364). The distinction between dignity and price corresponds to the dis-
tinction between ends that consist in possible results of action and ends
that are “self-existent.” The former, Velleman says, are objects of prefer-
ence and choice and are comparative. The latter are not produced by
action, and their value does not serve as grounds for comparing them with
alternatives but as grounds for revering or respecting them as they already
are. This value is incomparable in that “it calls for a response to the object
[that has this value] in itself, not in comparison with others” (p. 364).
Love, then, is a response to (as a result of being aware of) the incompa-
rable value possessed by a person in virtue of the person being a self-­
existent end.
Velleman further proposes that love is an arresting awareness of such
value. It is so in that, in responding to the incomparable value of a person
in the manner constitutive of love, our defenses against being emotionally
affected by the other are lifted (pp. 361, 366). Velleman explains that
conceiving of love as a response to a person’s rational nature may seem
odd if “rational nature” is taken to denote the intellect. But rational
nature, he says, is not the intellect. Rather, it is a capacity of valuation: “a
capacity to care about things in that reflective way which is distinctive of
self-conscious creatures like us” (p. 366). We are invited to think of a per-
son’s rational nature as his core of reflective concern (pp. 366–367). What
we respond to, then, in loving a person, is the value that the person has in
virtue of being a person. This value “inheres” in the capacity persons have
to appreciate the value of self-existent ends or, in other words, the capacity
 8 LOVE AND FREE AGENCY 155

persons have for loving others. So according to Velleman, “what we

respond to, in loving people, is their capacity to love: it’s just another way
of saying that what our hearts respond to is another heart” (p. 365). Since,
in loving another, we respond to their capacity to love us, we suspend our
emotional defenses against them:

[L]ove for others is possible when we find in them a capacity for valuation
like ours, which can be constrained by respect for ours, and which therefore
makes our emotional defenses against them feel unnecessary. That’s why our
capacity for valuation, when facing instances of itself, feels able to respond
in the manner constitutive of love, by suspending our emotional defenses.
Love, like respect, is the heart’s response to the realization that it is not
alone. (Velleman 1999: 366, note omitted)

Proponents of yet other competing accounts emphasize that part of the

complex state or condition of love is characteristically trust between the
lover and the beloved. Laurence Thomas, for example, claims that a distin-
guishing mark of friendship—and by extension, love—is the bond of
mutual trust between friends. This, he thinks, “is cemented by equal self-­
disclosure and for that very reason, is a sign of the very special regard
which each has for the other” (Thomas 1987: 217). Dean Cocking and
Jeanette Kennett take issue with Thomas’ proposal that self-disclosure—
the confiding of private or intimate information—between friends cements
bonds of mutual trust. But they agree that trust and intimacy are central
to friendship (Cocking and Kennett 1998).
Armed with these disparate proposed accounts of love’s nature, we may
attempt to galvanize the view that love is historical in the same fashion in
which many have tried to show that responsibility is historical: invoke
appropriate radical reversal cases. Preliminarily, keep in mind that nothing
in principle precludes manipulators of the sort we find in One Bad Day
from (1) implanting into unsuspecting agents desires of various sorts
including those (if any) fundamental to love; (2) making such agents emo-
tionally vulnerable to others or seeing to it that they disarm their emo-
tional defenses against others; (3) ensuring that such agents trust others;
or (4) duplicating in such agents the complex volitional structure of the
sort Frankfurt theorizes is central to love. Briefly, pick the feature that you
deem central to love. It looks as though any such feature can be covertly
reproduced in unsuspecting agents.
156 I. HAJI

Would an aptly developed radical reversal case affect our pertinent

judgments of love? Let’s see. To develop such a case, first, attend more
carefully to the notion of an agent’s values. An agent values something at
a time if, and only, if at that time she desires and believes it to be good
(Mele 2019: 14–15). Values, as psychological states, have both these
dimensions. An action expresses a value only if this value (or its neural real-
izer) plays an appropriate causal role in its production. Now, for the case,
assume that a cluster of values, VLuv, is fundamental to love. Romeo, but
not Romello, loves Juliet. Having examined what makes Romeo tick, the
manipulators implant in Romello Romeo’s VLuv values, and make other
necessary changes in Romello so that he is relevantly just like Romeo in
One Lovely Day—he now (supposedly) loves Juliet. By late evening,
Romello reverts to his former self, showing little interest in Juliet. During
the day, does Romello indeed love Juliet? One option is that he does not
because the provenance of VLuv matters. Since Romello does not acquire,
maintain, or endorse (in some suitable sense of “endorse”) VLuv on the
basis of his pre-manipulated values for which he was largely responsible—
since he is not the “ultimate originator” of VLuv—he does not love Juliet.
In favor of this option, the judgment that manipulated Sally (M-Sally)
is not morally blameworthy for killing George in One Bad Day is highly
plausible. However, assuming that it is morally wrong for pre-manipulated
Sally to kill George, it is not credible that, in virtue of being manipulated,
it is no longer wrong for M-Sally to kill George. Elsewhere, to account
partially for the asymmetry in these judgments—the engineered change in
Sally’s values affects our appraisal of moral responsibility but not of moral
obligation—I have suggested that appraisals of obligation are, first and
foremost, appraisals of what agents do—they are “act-focused,” rather
than appraisals of the agents themselves—they are not “agent-focused.”
With responsibility, or agent-focused appraisals more generally, the agent
is of paramount significance; her desires, beliefs, and values—her mental
repertoire—must be “unequivocally attributable” to her because she is the
primary object of evaluation (Frankfurt 2002: 27). This is not so with act-­
focused assessments of obligation. Imagine that a button in a hotel suite
in New York must be pressed to save millions of lives in Asia; it is morally
obligatory, at some time, for someone that she presses the button at this
time, no matter who this person is and no matter what her psychology—it
could be Michael, Bob, or Dana, however psychologically different they
may be. Roughly, the idea is that evaluations of obligation pay relatively
little heed to the sort of person you are, whereas those of responsibility do.
 8 LOVE AND FREE AGENCY 157

Perhaps, love is pertinently like responsibility and unlike obligation in this

respect. The agent who loves is deeply implicated in a way that, owing to
her profound “investment” in love, it is reasonable to judge that in radical
reversal cases such as One Lovely Day, the manipulated agent, like Romello,
does not love Juliet.
On the competing option—Romello does love Juliet in One Lovely
Day—one might take an internalist’s stance on love, analogous to the
stance Frankfurt takes on responsibility.

If someone does something because he wants to do it, and if he has no res-

ervations about that desire but is wholeheartedly behind it, then—so far as
his moral responsibility for doing it is concerned—it really does not matter
how he got that way. One further requirement must be added…: the per-
son’s desires and attitudes have to be relatively well integrated into his gen-
eral psychic condition. Otherwise they are not genuinely his…. As long as
their interrelations imply that they are unequivocally attributable to him….
it makes no difference—so far as evaluating his moral responsibility is con-
cerned—how he came to have them. (2002: 27)

One may opt for the view that just as the sort of manipulation in One
Bad Day fails to let Sally off the hook—one is a staunch responsibility
internalist—so the sort of homologous manipulation in One Lovely Day
fails to undermine the judgment that Romello does, indeed, love Juliet.6 I
have leanings toward the first option. I leave it to you, the reader, to
decide which of the two options you find more tenable.
We may now unearth an important first result. Even if the second
option has the upper hand, it is highly credible to regard Romello’s love
in One Lovely Day as forced, artificial, or ersatz. We value love in that we
are favorably disposed to and judge or believe it to be good. Assuming
Romello loves Juliet in One Lovely Day, this sort of ersatz love isn’t the
sort of love we value. It is “free love” and not the kind of contrived love
we find in One Lovely Day that we cherish.
Agency can be manifest, for instance, in performing intentional actions,
reacting with feelings or emotions, or expressing love. Radical reversal
cases bring into stark relief the value of autonomous agency. We don’t
esteem being the sort of agent whose intentional actions are solely the
product of covert manipulation, roughly, because our agency is hijacked.

6
Other responsibility internalists or those with leanings toward responsibility internalism
include Double (1991) and Watson (1999).
158 I. HAJI

Analogously, if we evince various feelings or express love only because we

have been aptly manipulated or conditioned, presumably we would not
value this manifestation of heteronomous agency.

8.4   Love’s Requirements and Acting from Love

One Lovely Day helps to reveal another somewhat related result having to
do not with love itself but with lovable behavior. The distinction between
acting in accordance with moral duty (or obligation) and acting from duty
is commonplace. Imagine that on some occasion you perform an act that
issues from your desire to amass long-term financial gain, moral consider-
ations do not figure at all in the proximal causal history of this action, and
you rightly believe that you act from prudential reasons. Suppose that
your act coincides with what you are morally required to do on this occa-
sion. Here, you have acted in accordance with, but not from, duty.
Similarly, we may distinguish between acting in accordance with or from
love. Several writers have plausibly proposed that there are requirements
and prohibitions from love’s standpoint. Frankfurt remarks:

It is characteristic of our experience of loving that when we love something,

there are certain things that we feel we must do. Love demands of us that we
support and advance the well-being of our beloved, as circumstances make
it possible and appropriate for us to do so; and it forbids us to injure our
beloved, or to neglect its interests. If we disregard these demands and pro-
hibitions, we feel that we are behaving badly—that we are betraying our
love. Now the grip and forcefulness of the requirements that love imposes
upon us resemble the forcefulness and grip of moral obligation. In cases of
both sorts—those involving love and those involving duty—it seems to us
that we are not free simply to do as we please or as we wish; love and duty
alike generate in us a sense that we have no choice but to do what they
require. (1999: 170)

Roger Lamb proposes that as a lover, you are obligated from love’s
standpoint to attend to requests of the beloved, help the beloved, be con-
cerned with the welfare of the beloved, and defend the trust that is partly
constitutive of your love (Lamb 1997: 28–29).
Acting from love is central to loving relationships, and to act from love,
one must be moved by love.7 If there is some sort of externalist constraint

7
See, for example, Pettit 1997: 155–156.
 8 LOVE AND FREE AGENCY 159

on love—if, in this way, love is historical—then acts that are tokens of act-
ing out of love will be indirectly historical too. Behavior that is loving
behavior must stem from the nuances or cares of love. When one loves
another, one is typically concerned for the other. The concern may express
itself in sundry ways, many behavioral. Insofar as the behavior that
expresses the concern is genuinely loving behavior, it seems that what is
done to manifest the concern must appropriately causally stem from love
and not, for example, from moral duty or prudence.
I previously proposed that in One Lovely Day, credibly, manipulated
Romello does not love Juliet. If so, his relevant behavior which would, in
the absence of manipulation, be lovable behavior, is not lovable behavior
because it is not motivated by love. Against this verdict, one may insist that
even in the throes of manipulation Romello does love Juliet, and hence his
relevant behavior is lovable behavior because it issues from love. However,
the love here is ersatz love, and acting from ersatz love is, again, not the
sort of lovable behavior we value.
A third lesson of the discussion on radical reversal cases and love is that
there is a sense in which love is fragile. Love has freedom or autonomy
presuppositions, and when these are undermined either love is under-
mined too or it isn’t the sort of love that is valued. Again, in virtue of
being aptly manipulated, and hence no longer pertinently free, either you
may think that Romello does not love Juliet in One Lovely Day, or if he
does, his ersatz love is not the sort of love we favor. As I’ll sketch in the
following section, sundry factors all beyond our control can affect free or
autonomous agency and, hence, affect love.

8.5   More on Love’s Fragility: Love and Moral

Sentiments or Attitudes
The intimate association between diverse emotions or attitudes and
healthy interpersonal relationships, including loving relationships, pro-
vides additional support for the thesis that love or at least loving relations
have freedom requirements. Justin Oakley writes:

There are several ways in which emotions may be construed as constituting

relationships of love and friendship. To begin with, the emotions we both
feel towards each other in a sense determine the form our relationship takes.
That is, our love or friendship for each other is embodied in our caring
about promoting each other’s welfare, our feeling sympathetic towards each
160 I. HAJI

other in regard to our respective problems, and our feeling angry and indig-
nant at injustices suffered by the other, to name only several. Further, emo-
tions may be thought of as constituting relationships of friendship and love
in as far as our mutual affection unifies and bestows a certain significance on
our joint activities. We see films and go on walks together out of love and
friendship, and many such activities, which might otherwise seem separate
and isolated, come to be seen as a complex whole in which our love and
friendship are manifested. (Oakley 1992: 58)

The general strand of reasoning to be developed (in this section) to

expose the freedom commitments of loving relationships is this: Integral
to these relationships are various emotions or attitudes. These attitudes
presuppose that we are responsible agents, which, in turn, requires that we
are relevantly free.
It has been plausibly proposed that remorse and guilt are significant to
interpersonal relationships partly because if someone were deprived of
them, she would be incapable of mending any relationships with people
whom she has wronged; and having done wrong, she would lack any
motivation to restore her own moral integrity and thus to develop morally.
But one’s guilt would be inapt if no one were ever free and, hence, not
morally responsible for anything. When one expresses gratitude to another
for what she has done, one can do so with the intention of developing a
relationship. However, gratitude is appropriate when the agent to whom
gratitude is owed is morally responsible for what she did; otherwise, grati-
tude would be misplaced. Similar things are true of forgiveness. “Well-­
founded” forgiveness presupposes that the person one forgives is
blameworthy for her untoward deed. Thus, each of these attitudes presup-
poses that we are relevantly free because each presupposes the existence of
responsibility.
What has been overlooked, especially with loving relationships, is that
the responsibility that attitudes like remorse, guilt, gratitude, and forgive-
ness implicate need not be moral. Rather, the praiseworthiness or blame-
worthiness may be praiseworthiness (“commendability”) or
blameworthiness (“censurability”) from love’s standpoint. To modify
Bernard Williams’ example, a spouse, saved by his wife who declares that
when she rescued him, she acted solely from moral duty and not love,
would surely be put off if he discovered the truth (Williams 1976: 214).
The wife may be morally praiseworthy but not commendable—not praise-
worthy from love’s standpoint—for saving her husband. The fundamental
 8 LOVE AND FREE AGENCY 161

suggestion is that, unlike what transpires in this case, typically, in loving

relationships the species of “normative responsibility” presupposed by
attitudes integral to these relationships is commendability or censurability.
Recall, we previously registered that there are duties (or obligations) and
prohibitions of love that may well be distinct from the duties or prohibi-
tions of morality. Assuming other requirements of moral responsibility are
satisfied, if you intentionally do something that is morally wrong (on some
views) or intentionally do something you nonculpably take to be morally
wrong (on competing views), you may well be deserving of moral blame.
Analogously, if you intentionally do something that love prohibits or you
nonculpably take love to prohibit, you may well be deserving of blame
from love’s standpoint. In brief, there are non-moral varieties of praisewor-
thiness and blameworthiness. A novice chess player who beats a grandmas-
ter may well be praiseworthy for her worthy moves but the praise here is
not moral. Analogously, you may do (or omit to do) something for which
you are blameworthy but the blame may be blame from love’s standpoint.
For example, you may be forgiven for ignoring a requirement of love that
is not a moral requirement when you are censurable—you are non-morally
but normatively blameworthy—for intentionally thwarting this require-
ment. In what would customarily be expected in a Williams-like story, the
wife would be commendable—again, non-morally but normatively praise-
worthy—for saving her husband. Or imagine that a mother visits her sick
child in hospital for no other reason than that she loves him and cares for
his well-being. The belief or thought that it is morally right or obligatory
for her to visit plays no role whatsoever in the etiology of this action. Any
such moral belief fails to enter into her deliberations (if she deliberates at
all) about whether to visit; nor does she entertain any moral belief in visit-
ing her child. The mother is not morally deserving of praise for visiting her
child. Or suppose the mother gives up one of her kidneys to her child who
would not otherwise survive. Assume that she acts from love and not
moral duty or any sense of moral concern. Then, again, the loving mother
seems not to be morally praiseworthy for giving the kidney. But she is
commendable; she is non-morally normatively praiseworthy. She gives up
her kidney, roughly, on the basis of the belief that this is what she ought,
from love’s perspective, to do. “Ought” in the previous sentence denotes
an obligation, or at least some prescriptive element like a duty or a deep
commitment, associated with acting from love that is somewhat analogous
162 I. HAJI

to what one takes to be one’s moral obligation when one acts in light of
the belief that one morally ought to do something.8
A few remarks on normative responsibility with attention primarily to
normative blameworthiness, are in order.9 As previously illustrated, there
are different species of normative blameworthiness. One can be morally,
love-wise, or skills-wise blameworthy for intentionally doing various
things. Normative blameworthiness is concerned principally with apprais-
ing an agent and only derivatively with appraising her behavior. Frequently
(but not always) it is associated with normative standards one takes to be
significant and calls upon to guide one’s conduct. These standards, broadly
construed, may include dictates of custom or tradition, or imperatives
deriving from projects or ideals of sizable importance in one’s life.
Furthermore, a set of dictates, ideals, or rules qualify as appropriate nor-
mative standards that “ground” normative responsibility provided they
guide and constrain behavior; they carry, in one’s life, a sort of normative
authority. An agent who accepts or endorses a set of normative standards
is motivated to act in accordance with them, believes that they provide
reasons for action, and is disposed to have (appropriate) pro or con feel-
ings or attitudes in conditions when these standards are implicated. Often,
when an agent is normatively blameworthy for something, she contra-
venes what she takes to be the dictates of the normative standards she
endorses. In virtue of such infringement it is frequently fitting when she is
normatively blameworthy for her to express apt attitudes (such as guilt or
remorse) and for others to adopt appropriate negative attitudes toward
her. Finally, there is no presumption that people generally endorse a single
set of ideals or standards that guide and constrain behavior across all
domains in their lives. With respect to certain concerns, one may act from
love, but regarding others, one may act from moral duty.10
If radical reversal cases like One Bad Day show that victimized Sally is
not morally blameworthy for killing George, then appropriately con-
structed analogous cases, such as One Lovely Day, should show that the

8
Lamb proposes that love involves being committed to the beloved where the sense of
“commitment” is a sense referring to our obligations as lovers (1997: 28). On Velleman’s
Kantian view of love, however, moral obligations and those of love do not come apart in
this way.
9
Much of what I say on normative blameworthiness will also apply, with appropriate
amendments, to normative praiseworthiness.
10
For more on the concept of normative blameworthiness, see Haji 1998, ch. 11; Haji and
Cuypers 2008: 108–111.
 8 LOVE AND FREE AGENCY 163

victimized character, M-Romello, is not censurable (or commendable) for

his pertinent deeds.
Summarizing, the view that “emotions may be construed as constitut-
ing relationships of love and friendship” exposes freedom commitments of
loving relationships because many of these emotions presuppose that we
are morally or non-morally responsible for our germane behavior, and
these varieties of responsibility have free agency commitments.

8.6   A Worthy Objection: Love’s Resilience

I’ve offered considerations for the view that love or lovable behavior, or
the sort of love or lovable behavior worthy of value, is free or autonomous
love or autonomous lovable behavior; lack of free agency compromises
love. In addition, freedom commitments also underwrite loving relations.
On a noteworthy and opposed view that Pereboom defends, which I now
briefly address, living without free will doesn’t hinder love. Owing to
space restrictions, I confine discussion to Pereboom’s views on the moral
emotions or attitudes (hereafter “attitudes”) central to salutary interper-
sonal relationships.
Construe responsibility skepticism as the view that moral responsibility
is non-existent in our (the actual) world because sources over which agents
in this world lack control ultimately produce their behavior. Not being the
ultimate originators of any of our behavior, we aren’t free and, so, aren’t
morally responsible for anything. Pereboom is a responsibility skeptic but
an optimistic one in that he argues that living without moral responsibility
is relatively inconsequential. It’s so because responsibility’s non-existence
leaves basically untouched, among several things important to our lives,
love and the moral attitudes essential to loving relations (Pereboom
2001, 2014).
What fundamentally undergirds Pereboom’s responsibility skepticism is
the following principle:

Primary Principle: Behavior produced by sources over which their agents

lack control, such as actions whose etiologies feature significantly intrusive
manipulation, is behavior for which they are not responsible. (Pereboom
2001: 4, 43)

Keeping this grounding principle in mind, it is fairly straightforward to

appreciate Pereboom’s skepticism regarding free agency. Take
164 I. HAJI

determinism to be the doctrine that all events are causally determined in

that they are the consequences or products of the laws of nature (whatever
they are) and facts of the world in the distant past.11 Indeterminism is the
denial of determinism. Either determinism or indeterminism is true at our
world. If the former, then we are not responsible for any of our choices or
actions because we are not the ultimate originators of them. Our conduct
is the product of the distant past and the laws, sources over which we have
no control. Pereboom further proposes that there is no relevant and prin-
cipled difference between an action resulting from responsibility-­
undermining manipulation (as in radical reversal cases like One Bad Day)
and an action with a mundane deterministic causal history. Since we are
not responsible for the pertinent actions that feature responsibility-­
subverting manipulation, given the “no difference” proposal, it follows
that we would not be responsible for any deterministically produced
actions. Suppose indeterminism is true at our world. Pereboom argues
that in scenarios involving indeterminism, just as in those involving deter-
minism, antecedents over which an agent lacks any control produce her
actions. Again, Pereboom’s verdict is that no relevant and principled dif-
ference can distinguish an action resulting from responsibility-­undermining
manipulation from an action with a more ordinary non-deterministic
causal history (Pereboom 2002: 478). He concludes that regardless of
whether determinism or indeterminism is true at our world, no one is
morally responsible for anything.12
We now have a serious concern regarding interpersonal relations. As we
briefly sketched, these relations (including loving relations) are intimately
associated with attitudes such as gratitude, forgiveness, or indignation, but
these attitudes assume that we are morally responsible agents. Responsibility
skepticism, thus, seemingly acutely endangers these relations.
Pereboom has a two-pronged response to this concern. He maintains
that, all things considered, some attitudes deemed vital to relationships do
more harm than good. For instance, he argues that indignation and moral
anger are not obviously required for good relationships (Pereboom 2001:
207–213). I set this insightful suggestion aside because it’s not relevant to

11
More rigorously, determinism is the thesis that at any instant exactly one possible future
is compatible with the state of the universe at that instant and the laws of nature.
12
Pereboom allows that if we were agent-causes—we as agents, as opposed to events
involving agents, are irreducible causes of some events—we could be responsible for some of
our conduct. But this isn’t borne out by our best scientific theories (Pereboom 2001: 69–88).
 8 LOVE AND FREE AGENCY 165

the ensuing discussion. In addition, Pereboom avers that the primary prin-
ciple leaves untouched attitudes such as gratitude and forgiveness or core
elements of them, which, he concurs, figure centrally in healthy
relationships.
Regarding the second prong, one of Pereboom’s principal defensive
maneuvers distills to this: (1) Various moral attitudes play key roles in ini-
tiating or maintaining relationships. For example, good friends or lovers
customarily forgive and forgiveness does indeed presuppose that the for-
given is blameworthy for her relevant behavior.13 (2) The primary princi-
ple undermines constituents of attitudes critical to relationships that
implicate responsibility. (3) However, responsibility’s non-existence leaves
unaffected core components of these attitudes or apt replacements for
them. (4) The impervious components or replacements can play the ger-
mane roles in relationships that the original attitudes do. (5) So, responsi-
bility skepticism does not debunk interpersonal relationships.
To illustrate, Pereboom claims that even if responsibility skepticism is
true, one may feel profound sorrow and regret on being the instrument of
wrongdoing despite believing that one was not in any way blameworthy.
He recommends that sorrow and regret can play the pertinent roles that
remorse and guilt typically do in relations. For example, sorrow and regret
may generate a repentant attitude and thus induce the agent not to per-
form her immoral action again; they may motivate the agent to make
amends by seeking to alleviate the suffering caused to others; and they
may help to heal the relation by impelling the agent to express misgiving
about her untoward behavior (Pereboom 2001: 205–206). So although
gratitude and guilt “would likely be theoretically irrational” for a respon-
sibility skeptic, these attitudes “have analogs that could play the same role
they typically have” (Pereboom 2001: 206).14
Pereboom’s second prong to secure interpersonal relationships in a
world devoid of responsibility is, however, suspect. The root of the worry
is that the primary principle casts its net exhaustively, enmeshing all behav-
ior including the choices or decisions anyone makes and the attitudes any-
one expresses. All these things originate in sources beyond anyone’s
control; so, no one is the ultimate originator of any of them. The import
of this primary principle’s implication for relationships is straightforward.

13
Pereboom is concerned with moral blameworthiness or moral responsibility generally.
I’ve proposed that the responsibility may be responsibility from love’s standpoint.
14
See also Pereboom 2014: 179.
166 I. HAJI

Imagine that Romeo and Juliet are aware that no one is ever morally or
non-morally responsible for anything. If Romeo expresses sorrow, an
alleged analog of guilt, on a particular occasion in a world without respon-
sibility because he believes he has wronged Juliet, the primary principle
implies that not only is Romeo not responsible for his expression of sor-
row, but this token of expression is essentially no different than a token of
sorrow he would have expressed had he been suitably manipulated. Simply
imagine a radical reversal case in which, owing fundamentally to the
manipulation, manipulated Romeo expresses sorrow unlike his pre-­
manipulated self. How effective would this token of sorrow be as a vehicle
to mending the relationship or as a motivator to restoring Romeo’s moral
integrity? Hardly at all, it appears. Aware that Romeo’s token of sorrow
might just as well have been the product of nefarious manipulation, why
should Juliet regard Romeo’s (unfree) expression of sorrow as “truly his
own” and, thus, as conducive to healing the wound? It is free or “authen-
tic” sorrow and not “forced” or ersatz sorrow that is beneficial to mend-
ing interpersonal relationships.
To amplify, just as we may distinguish between “authentic” springs of
action, such as desires that are “truly our own,” and inauthentic springs of
action, so we may distinguish between authentic attitudes, authentic sor-
row, for instance, and inauthentic attitudes. In One Bad Day, Sally’s engi-
neered-­in Chuck-like values are not “truly her own” or authentic. If she
were to express joy on successfully executing George, and this joy were to
issue from her implanted values, her token expression of joy would, simi-
larly, not be authentic. Presumably, the responsibility skeptic like Pereboom
accepts the condition that a desire or an attitude is authentic only if it does
not ultimately derive from sources over which the agent lacks any control.
The original concern responsibility skepticism generates for some attitudes
pivotal to healthy interpersonal relationships is that they are inappropriate
in a no-responsibility world owing to presupposing responsibility, some-
thing non-existent in such worlds. The responsibility skeptic proposes that
the role these threatened attitudes play in relationships can be assumed by
other attitudes that are on sure footing in such worlds. However, if there
is a legitimate distinction between, for example, free or authentic and
unfree or inauthentic sorrow, and all sorrow in a no-responsibility world
is, as the responsibility skeptic is committed to conceding, unfree or inau-
thentic, then the original concern of responsibility skepticism regarding
attitudes such as guilt resurfaces with the proposed replacements.
 8 LOVE AND FREE AGENCY 167

One may rejoin that the point of sorrow’s replacing guilt in Pereboom’s
discussion is that sorrow is unlike guilt, presumably, because it’s a feeling
that is not subject to the inauthentic/authentic divide. One could argue
that genuine sorrow just happens to us—like a more sophisticated version
of pain—and even if sorrow is implanted in us by malevolent manipula-
tors, we’re still genuinely sad.15 However, this objection overlooks a cru-
cial and contentious factor in Pereboom’s strategy to secure sorrow as a
replacement for guilt: even engineered-in sorrow can play the same role as
guilt does—for instance, healing a relationship—in our day-to-day interac-
tions. Imagine that your world is deterministic; Seth unjustifiably harms
you and then expresses sorrow. The event, Seth’s expressing sorrow, is deter-
ministically produced and, hence, in Pereboom’s view not relevantly dif-
ferent from such an event that Seth would have brought about but only
because of the kind of manipulation manifest in One Bad Day. Knowing
this, how effective would such an expression of sorrow be in mending the
relationship? If you feel guilt only because you’ve been manipulated to feel
it and the other party knows this, such guilt is not going to help set mat-
ters right. Why should things be any different if you’re manipulated to feel
sorrow instead of guilt?16

8.7   Final Thoughts

You may wonder why we should care about Romello’s plight when it has
nothing to do with our own. He’s a puppet on the manipulators’ strings,
we’re not! Exposing the freedom presuppositions of love should give us
reason for pause. Pereboom’s argument for free will skepticism may con-
vince you that we are not free agents. Or other arguments for free will
skepticism may sway you.17 Or, even if you think that determinism and
indeterminism can both accommodate free will, you may be persuaded
that various factors that are out of our control—such as our genetic con-
stitution and the environment in which we were brought up—leave us
substantially unfree. Indeed, I believe that a large swath of our conduct
that we normally think is within our control really isn’t. For instance, what
choices we are able to make is dependent upon what motivations we have,

15
I thank Simon Cushing for this objection.
16
There is another concern with the objection that I here simply flag. Sorrow is relevantly
unlike pain in that you can unjustifiably express sorrow.
17
See, for example, Smilansky 2000; Strawson 1986.
168 I. HAJI

but what motivations we have is very often beyond our control and so
subject to freedom-undermining luck (see Haji 2016). If we’re substan-
tially unfree, then love will be compromised because love is, oh so fragile!18

References
Cocking, Dean, and Jeanette Kennett. 1998. Friendship and the Self. Ethics
108: 502–527.
Double, Richard. 1991. The Non-Reality of Free Will. New York: Oxford
University Press.
Frankfurt, Harry G. 1999. On Caring. In Necessity, Volition, and Love, 155–180.
Cambridge: Cambridge University Press.
———. 2002. Reply to John Martin Fischer. In Contours of Agency: Essays on
Themes from Harry Frankfurt, ed. S. Buss and L. Overton, 27–31. Cambridge,
MA: MIT Press.
Green, O.H. 1997. Is Love an Emotion? In Love Analyzed, ed. Roger E. Lamb,
209–224. Boulder: Westview.
Haji, Ishtiyaque. 1998. Moral Appraisability: Puzzles, Proposals, and Perplexities.
New York: Oxford University Press.
———. 2009. Incompatibilism’s Allure: Principal Arguments for Incompatibilism.
Peterborough, ON: Broadview Press.
———. 2013. Historicism, Non-Historicism, or a Mix? Journal of Ethics
17: 185–204.
———. 2016. Luck’s Mischief. New York: Oxford University Press.
Haji, Ishtiyaque, and Stefaan Cuypers. 2008. Moral Responsibility, Authenticity,
and Education. New York: Routledge.
Kolodny, Niko. 2003. Love as Valuing a Relationship. Philosophical Review
112: 135–189.
Lamb, Roger E. 1997. Love and Rationality. In Love Analyzed, ed. R.E. Lamb,
23–47. Boulder: Westview Press.
Mele, Alfred. 2006. Free Will and Luck. New York: Oxford University Press.
———. 2016. Moral Responsibility: Radical Reversals and Original Design.
Journal of Ethics 20: 69–82.
———. 2019. Manipulated Agents: A Window to Moral Responsibility. New York:
Oxford University Press.
Oakley, Justin. 1992. Morality and the Emotions. New York: Routledge.
Pereboom, Derk. 2001. Living Without Free Will. Cambridge: Cambridge
University Press.

18
This paper was completed during my tenure of a 2017–2021 Social Sciences and
Humanities Research (SSHRC) grant. I thank this granting agency for its support. I’m most
grateful to Simon Cushing for his comments and suggestions.
 8 LOVE AND FREE AGENCY 169

———. 2002. Living Without Free Will: The Case for Hard Incompatibilism. In
The Oxford Handbook of Free Will, ed. Robert Kane, 477–488. New York:
Oxford University Press.
———. 2014. Free Will, Agency, and Meaning in Life. New York: Oxford
University Press.
Pettit, Philip. 1997. Love and Its Place in Moral Discourse. In Love Analyzed, ed.
R.E. Lamb, 153–163. Boulder: Westview Press.
Smilansky, Saul. 2000. Free Will and Illusion. Oxford: Clarendon Press.
Strawson, Galen. 1986. Freedom and Belief. Oxford: Oxford University Press.
Thomas, Lawrence. 1987. Friendship. Synthese 72: 217–236.
Velleman, David. 1999. Love as a Moral Emotion. Ethics 109: 338–374.
Watson, Gary. 1999. Soft Libertarianism and Hard Compatibilism. Journal of
Ethics 3: 353–368.
Williams, Bernard. 1976. Persons, Character, and Morality. In The Identities of
Persons, ed. A. Oksenberg Rorty, 197–215. Berkeley: University of
California Press.
 CHAPTER 9

Sentimental Reasons

Edgar Phillips

9.1   Introduction
Among the things we do, some we do for reasons and some we don’t.
When we act intentionally, for instance, we normally have some reason for
doing what we are doing. We believe for reasons and choose for reasons.
We do not, at least not in the same sense, sweat, feel tired, or digest our
food for reasons. Other examples are somewhat less easy to categorize.
For example, it might not seem as immediately clear whether we feel emo-
tions for reasons; there is, after all, a long tradition of thinking of the emo-
tions as interfering with our ability to act and think rationally or in line
with reason. While emotions certainly can interfere with our ability to
remain rational, though (we all know what it’s like for our emotions to get
the better of us), this doesn’t necessarily mean we have no reasons for our
emotions. And on reflection, it is quite natural to think that we do: when
you feel angry or relieved, for instance, there will be, except perhaps in
marginal cases, something that made you angry, or something that you
feel angry or relieved about or by. It is quite natural here to use the

E. Phillips (*)
Institut Jean Nicod at the Ecole Normale Supérieure, Paris, France

© The Author(s), under exclusive license to Springer Nature 171

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_9
172 E. PHILLIPS

language of reasons: your reason for feeling angry is that your friend
betrayed you; the reason for your relief is that the operation is finally over.1
Things seem even less clear if we turn to consider whether we love for
reasons.2 While a number of philosophers have claimed that we do not
love for reasons, many insist that we do. Interestingly, even among those
who hold that we love for reasons, there is remarkable disagreement as to
what sorts of things those reasons are, so much so that it seems sensible to
consider carefully just what it is that the disagreement is about: what
should an account of ‘reasons for love’ seek to account for? The aim of this
chapter is to try to reach a degree of clarity about this question and in
doing so to shed some light on debates about reasons for love. To this
end, I will first, in the next section, give a brief overview of some leading
views about reasons for love, showing how different kinds of consider-
ations seem to pull us toward very different views. I will then introduce a
distinction commonly made in discussions of reasons for action, between
three different roles that reasons can play, or between three different inter-
ests that commonly figure in our talk about ‘reasons’. As I’ll explain, it is
attractive to see these interests or roles as coinciding in a certain way when
someone does, thinks, or feels something ‘for a reason’. One way to
understand the nature of the debate about reasons for love, I will suggest,
is that many of those involved in this debate assume that the same kind of
coincidence of interests must be present in the case of love if indeed we
love ‘for reasons’. In the end, I’ll suggest that we might make better sense
of love if we were to drop this assumption.

9.2   Reasons for Love

Do we love for reasons? It can be tempting to say that we don’t. In a
romantic mood, the suggestion that love is based on reasons can look
absurdly or even creepily high-minded, cold-hearted, and calculating.
Love, after all, is the paradigmatic passion. Compared with the actions we

1
There is, of course, much more to be said about the relationship between emotions,
reasons, and rationality. For a seminal discussion, see de Sousa (1987). See also Deonna and
Teroni (2012).
2
If love is an emotion, the point here is that things are less clear in the case of love than
they are in the case of certain other emotions. But perhaps love is not itself an emotion, even
though it is intimately connected with emotions. Either way, I will argue later that there are
significant differences between love and emotions like anger and relief, such that love
deserves separate treatment.
 9 SENTIMENTAL REASONS 173

take and attitudes we hold for reasons, love comes from somewhere deeper
in our soul. We don’t decide to love: we fall in love; love overcomes us,
sweeps us off our feet, carries us away; love is, in Nick Zangwill’s phrase,
‘gloriously arational’ (Zangwill 2013). On the other hand, though, love
seems closer to those emotions that we do (plausibly) feel for reasons than
it is to clearer examples of things we do but not for any reason, such as
sweating, digesting, or feeling tired. Love, from the lover’s perspective,
feels like a ‘fitting’ response to the beloved, in something like the way that
anger feels like a fitting response to being wronged, or fear a fitting
response to imminent danger. Moreover, loving someone seems to involve
our valuing or caring about them, and the things we value or care about
are not, from our point of view, arbitrary or selected merely by chance:
they seem to us worth valuing, worth caring about. What could make
something worth valuing except a reason to value it?
So there is some pressure to say that we love for reasons. Trying to say
anything much about what sorts of things those reasons are, though,
proves to be rather more difficult—more difficult, notably, than in the case
of our other examples of ‘reasons-responsive’ phenomena. Reasons for
believing seem clearly enough to be concerned with the truth of the things
we might be inclined to believe. The fact that the streets are wet, for
instance, might be a reason to believe that it has rained, because it suggests
that it has in fact rained. Reasons for feeling emotions generally concern
whether the ‘object’ of the emotion (the thing, event, or person toward
which the emotion is felt) exemplifies what is called the formal object of the
emotion: danger or fearsomeness for fear, wrongfulness or insult for anger,
and so on (see Deonna and Teroni 2012 for a detailed exposition of this
idea). So, for instance, my reason for feeling angry at you might be that
you betrayed my trust, this being a case of you wronging me. Reasons for
action appear to be more diverse and there is a greater degree of disagree-
ment over what unites them, but we can at least say that they broadly
relate to the point of taking one or another course of action.3 Your reason
for taking your umbrella when you go out, for instance, might be that it

3
One way to explicate this idea, for example, is to say that reasons for action relate to the
good at which the action aims: the good in question gives the action its point (e.g. Raz
1999). A different approach appeals to the desires of the agent: the point of acting is to sat-
isfy one’s desires (e.g. Schroeder 2007). Perhaps the right account combines both ideas
somehow (e.g. Chang 2011), or appeals to something else such as norms or rules or rational-
ity (e.g. Korsgaard 1996). Since our present concern is with love, not action, there is thank-
fully no need for us to take a stand on this difficult issue here.
174 E. PHILLIPS

seems likely that the rain will return and that the umbrella will help keep
you dry; your reason for apologizing to me might be that you did me
wrong and apologizing will be the first step to mending our damaged
relationship. If we suppose that there are indeed reasons for love, though,
what sorts of things might they be?
An initially appealing strategy would be to analogize love to emotions
like fear or regret and to say that reasons for love concern whether the
actual or potential ‘object’ of love—the actual or potential beloved—
instances the ‘formal object’ of love. As the formal object of fear is the
fearsome, and that of regret is the regrettable, so the formal object of love,
we might suggest, is the lovable (compare Naar 2017b). So, reasons for
love might be facts about a person that suggest that that person is lovable,
or perhaps properties of the person in virtue of which they are lovable.
However, while this suggestion makes for a nice linguistic consistency
with emotions like fear and regret, it doesn’t fit very well with how we
ordinarily think about love, at least if we are using ‘lovable’ in its everyday
sense. To be lovable in that sense is, roughly, to be easy to love. In particu-
lar, certain qualities, perhaps including things like charm, a gentle wit,
kindness, and innocence might tend to make one who possesses them lov-
able. Surely, though, lacking such qualities doesn’t mean that nobody has
any reason to love you. Most people, I suspect, will have someone—a
friend, a family member, perhaps even a romantic partner—whom they
love and care about despite that person’s not being particularly lovable.
Moreover, a complete stranger might be exceedingly lovable without this
giving me any particular reason to love them.
The appeal to ‘lovableness’, then, seems not to give a very satisfactory
account of what reasons for love consist in. We might, however, think that
this first attempt does get something importantly right, namely, that rea-
sons for love consist in personal qualities or properties of the beloved.
Sometimes the qualities of a person that give us reason to love them will
be ones that make them lovable, but they might also often include quali-
ties that would not so naturally fall under this heading, such as, perhaps,
cleverness, resilience, or bravery. Even among authors who agree on the
basic idea that reasons for love are personal qualities, though, there is dis-
agreement as to what the relevant qualities have in common. Neil Delaney,
for instance, argues that we want to be loved for properties that we take to
be central to our conception of ourselves (Delaney 1996; also compare
Keller 2000), while Kate Abramson and Adam Leite hold that reasons for
love consist in ‘morally laudable’ qualities of the beloved’s character
 9 SENTIMENTAL REASONS 175

(Abramson and Leite 2011). While these might sometimes coincide, they
can obviously come apart. Most of us do not conceive of ourselves exclu-
sively in terms of our morally laudable qualities.
The idea that personal qualities are reasons for love looks appealing
because it seems both to capture the idea of love as a response to another
person as such (as opposed to, say, anger, which, while typically directed at
a person, is usually a response to something that they have done rather
than to who they are or what they are like) and, since different people have
different qualities, to offer the beginnings of an account of why we love
some people and not others. Yet some of the most compelling criticisms of
the so-called quality view are in fact based on the apparent shallowness of
the account that it offers of the latter phenomenon, the ‘selectivity’ of
love. Love, as we ordinarily think of it, involves a deep attachment to par-
ticular individuals. A series of objections to the quality view have been
taken to show that it cannot make sense of this aspect of love:

• Universality: If my reasons for loving you are qualities you have,

should anyone else who is aware of those qualities love you too and
in the same way?
• Promiscuity: Should I love, in the same way, anyone else who has the
same qualities?
• Trading up: If someone else has the same qualities to a greater
degree, should I love them instead, or more?
• Inconstancy: If you lose the relevant qualities, should I stop loving
you? (adapted from Setiya 2014: 255).

It seems very plausible that the answer to each question should be neg-
ative. This imposes an explanatory challenge upon a proponent of the
quality view. One way in which they might attempt to meet this challenge
is to say that the reasons for loving someone are not ‘requiring’ or ‘maxi-
mizing’ (Abramson and Leite 2011; Jollimore 2011; Setiya 2014). To say
that reasons for love are not requiring would mean that they can make a
person fitting or ‘eligible’ to love without obliging anyone to love them.
To say that they are not maximizing would mean that from the fact that
there is more reason to love person B than person A, it does not follow
that you ought to love person B rather than person A. If the reasons for
love are non-requiring and non-maximizing, this allows us to respond to
the aforementioned objections in the following ways. First, if my love for
you is justified, it may follow that anyone else who is aware of the qualities
176 E. PHILLIPS

that justify my love has sufficient reason to love you and could in principle
love you justifiably. However, it does not follow that they must love you:
the qualities in question make it appropriate for them to love you, without
making it inappropriate for them not to love you. Second, if the qualities
that justify my love for you do not require me to love you, then I am also
not required to love anyone else with the same qualities. Third, since the
reasons for love are not maximizing, I am not required to love anyone
who possesses the same qualities to a greater degree or extent.
This addresses the objections from Universality, Promiscuity, and
Trading Up. What about Inconstancy? One possible deflation of the chal-
lenge is this (compare Jollimore 2011): everyone, or just about everyone,
has some good qualities. If love is not maximizing, perhaps it does not
take very much for love to be justified, and it is enough that the actual or
potential beloved has some good qualities. Moreover, it is part of loving
someone, at least normally, to see the good in them, so that once you
already love someone you will not easily take them to have no good quali-
ties. If so, then the kind of change of character necessary for one to lose
one’s justification for loving someone might have to be really quite
extreme: they might have to become a kind of monster. In that case, per-
haps one really ought to stop loving them.
Even if the challenges from Universality, Promiscuity, Trading Up, and
Inconstancy can be addressed in these ways, though, there is a further and
more difficult challenge to the quality view. This is presented by Niko
Kolodny as the problem of Nonsubstitutability:

If Jane’s qualities are my reasons for loving her, then they are equally reasons
for my loving anyone else with the same qualities. Insofar as my love for Jane
is responsive to its reasons, therefore, it ought to accept anyone with the
same qualities as a substitute. But an attitude that would accept just as well
any Doppelgänger … that happened along would scarcely count as love.
(Kolodny 2003: 140–141)

Making the reasons for love non-requiring and non-maximizing does

not help here. The point about nonsubstitutability that poses the problem
for the quality view is not just that I am not required to love the
Doppelgänger; it’s that a willingness to accept the Doppelgänger as a sub-
stitute or replacement for Jane would show that I did not really love her at
all. The point of the nonsubstitutability objection is that love (as Kraut
1987 emphasizes) picks out the beloved as a particular individual. Personal
 9 SENTIMENTAL REASONS 177

qualities, being the kind of thing that can in principle be shared by mul-
tiple people, seem insufficiently ‘particular’ to explain this aspect of love.
Different authors have sought to resolve this difficulty in different ways,
but a common strategy is to appeal to historical or relational factors to
explain why one person’s love for another picks out that other as a particu-
lar individual. An influential version of this approach is developed by
Kolodny (2003), who argues that your reason for loving someone consists
in the fact that you have a valuable relationship with them. Because your
relationship with another person is a relationship with them and no one
else, this explains the nonsubstitutability of the beloved in an attractively
straightforward manner. Moreover, there seems to be something right in
the idea that your love for another person has something important to do
with your personal history with them, particularly when that history is
relatively long and the love is deep and abiding. If you ask yourself why
you love your partner, you might, in line with the quality view, think about
all the wonderful things about them. Equally, though, you might, as the
‘relationship view’ would suggest, think about all the times you have had
together, the things you have done for and with one another, your strug-
gles and triumphs and adventures, and so on. The relationship view neatly
explains why such past events should seem significant.
With that being said, the relationship view still strikes many as implau-
sible in important respects (for criticisms of Kolodny’s account, see, e.g.,
Smuts 2014; Setiya 2014; Na’aman 2015; Protasi 2016). A central con-
cern is that the view loses the idea that love is in the first instance a response
to another individual, not to one’s relationship with that individual. Other
authors have sought to preserve this idea while taking advantage of the
relational aspect of Kolodny’s account, suggesting that the presence of a
relationship with the other person is a kind of background condition, so
that other considerations—in particular, considerations about the other
person’s qualities or character—are the lover’s reasons for loving them,
but that they only count as such in a suitable ‘relational context’ (Abramson
and Leite 2011; Naar 2017b, forthcoming). When you think about your
history with your partner, on this view, what matter are the qualities of
character that they expressed in those past interactions with you: the ways
in which they showed you their kindness, generosity, courage, and the like.
Again, though, there are problems. How does this kind of view explain
parents’ love for their children, for instance? It’s doubtful whether infants
in particular have any morally laudable qualities of character, and certainly
they will not have had much opportunity to express any such character in
178 E. PHILLIPS

interactions with their parents, yet it goes without saying that parents nor-
mally feel profound love for their infant children. Indeed, familial love can
pose problems even when the beloved is an adult. People who are seriously
lacking in good qualities of character are often still loved by their parents
or siblings. Perhaps there is something pathological or irrational about a
parent’s love for their amoral or vicious offspring—but perhaps not. We
often speak of ‘unconditional love’ as something admirable, not some-
thing to be condemned.
We might be encouraged, by all these difficulties, to try a radically dif-
ferent approach. Some authors, for instance, have argued that the reasons
for love are minimal and universal: that someone’s mere humanity (Setiya
2014), or perhaps their personhood (Velleman 1999), is sufficient reason
for loving them. This has the appealing implication that it is never a mis-
take, never inappropriate or unfitting, to love another human being. There
is nothing wrong with loving a vicious family member—or indeed anyone
else for that matter. But while this sounds like a nice idea, it lacks any obvi-
ous way of accounting for the selectivity of love. It seems from my point
of view that I have special reasons for loving my partner of many years,
reasons I would not have for loving anyone else and especially not some
random stranger. Yet my partner and the stranger are just as much human
beings, just as much persons.4

9.3   Disentangling ‘Reasons’: Explanation,

Understanding, Justification
In seeking a unified account of reasons for love, we seem to be pulled in
different directions. Moreover, we seem to be pulled in these different
directions by different kinds of considerations. The ideas of love as a
response to another person, as involving a kind of appreciation of that
person, and the idea that we want to be loved for who we are suggest a
view of reasons for love as personal qualities of the beloved. What I’ve
called the selectivity and particularity of love, its character as an attach-
ment to a particular individual as such, emphasize the significance of his-
torical and relational factors. Both personal qualities and historical–relational
factors seem to be relevant in explaining why we love the specific people we

4
Velleman (1999: 370ff.) makes an attempt at squaring this circle. Setiya (2014) avoids it
by allowing that while someone’s humanity is sufficient reason to love them, having a rela-
tionship with them can provide a further, more forceful reason to love them.
 9 SENTIMENTAL REASONS 179

love. And a view that emphasizes either one of these factors in character-
izing reasons for love seems to risk conflict with an appealing ethical ideal
according to which no one is unworthy of love and that it is never inap-
propriate to love someone.
The multifariousness of considerations at play here makes it difficult to
arbitrate between different theories. If we are pulled in different directions
by different kinds of considerations, which should we prioritize? This, in
turn, points us toward a more basic question: What exactly is it that these
theories are disagreeing about? That is, what is a theory of ‘reasons for
love’ supposed to explain? One way we might try to gain some clarity here
is to think about the more general notions of ‘a reason’ and of doing
things ‘for reasons’. As we noted earlier, there are phenomena other than
love to which these notions are more obviously applicable—notably, belief
and intentional action. Questions about reasons for belief and action and
what it is to believe or act for a reason have also received rather more
philosophical attention than the corresponding questions about love. By
looking at some distinctions and ideas developed in discussions of reasons
for action and belief, then, we might hope to bring some more clarity to
our questions about reasons for love.
It is commonplace in the philosophy of action to make a distinction
between kinds of reasons, in particular between justifying (or ‘norma-
tive’), explanatory, and motivating reasons. Another way to put it, which
allows for the possibility that one reason might justify, explain, and moti-
vate, is that these are different roles that reasons can play (Alvarez 2010).
Even less committally, we might just distinguish different interests that can
be operative in our talk about reasons: sometimes we are concerned with
what we ought to do or with a person’s justification for what they did;
sometimes we are interested in why someone did what they did; sometimes
we are interested in what motivated the person to do what they did, or on
what grounds or basis they did it (compare Fogal 2018).
To illustrate the distinction, let’s take a concrete example. I knock your
favorite cup off the table. The cup smashes on the floor, leaving tea and
shards of china all over the parquet. You might well want to know why I
did this. Anything that accurately answers that question (hence that
explains my action), we can call an explanatory reason. Suppose I knocked
the cup off the table because I saw a face at the window and it made me
jump. In this case, the fact that I saw a face at the window is an explanatory
reason for my action—it’s a reason why I knocked the cup off the table. It
needn’t be the only such reason: we could equally cite the fact that the face
180 E. PHILLIPS

startled me, that I am a jumpy person, that the cup was too close to the
edge of the table, and so on. In general, a whole host of factors will be
relevant for the purposes of explaining any given action. Which constitute
the best or most relevant explanation will depend on various factors—
most obviously, on what is already known or assumed in the context of
explanation.
Now, consider another version of the example where a different expla-
nation is available. Suppose I knock the cup off the table in order to get
back at you for some perceived slight. Here we can explain my action in a
special way: by giving my reasons for knocking the cup off the table. Maybe
I did it because I wanted to get my own back at you, or because you
offended me, or because I thought you deserved to have your favorite cup
ruined. These are still ‘explanatory reasons’ in a perfectly good sense, but
they are also my reasons in a sense that the explanatory reasons considered
above were not. When we say that I knocked the cup off the table because
you offended me, or because I wanted to upset you, we explain my action,
but we explain it specifically by showing what point there was, from my
point of view, in taking that course of action, thus revealing my purpose or
intention in doing what I did. These kinds of explanations make sense of
my action by showing what motivated me so to act. Hence, the kinds of
factors cited in such explanations are commonly called motivating reasons.
Something like the distinction between explanation in general and a
narrower kind of explanation in terms of the person’s own reasons seems
to have application beyond the case of action, including to phenomena—
in particular to certain sorts of psychological states or attitudes—with
regard to which the language of ‘motivation’ seems less apt. So, for
instance, if you ask me why my friend believes that 5G makes people sick,
‘Because they read it online’ seems to give something like the kind of
explanation we are interested in, while ‘Because they’re a paranoid con-
spiracy theorist’ doesn’t. The former tells us something about the basis on
which my friend believes this claim and so tells us something about their
reasons for believing it, even though it doesn’t identify a ‘motivation’ for
believing it.5 Similarly, the density and speed of traffic might explain my

5
Beliefs can sometimes be motivated. People do sometimes engage in motivated reasoning
or wishful thinking, believing things that they want to believe because they want to believe
them. We tend to think of such beliefs as irrational, and it seems that believing in this way
involves some degree of self-deception, with the believer convincing themselves that they
actually have good grounds for believing as they do. While this phenomenon may be more
 9 SENTIMENTAL REASONS 181

fear of crossing the road by giving my reason for feeling afraid. Again,
though, the traffic does not motivate me to feel frightened; it simply
frightens me. It’s plausible, then, that we might make a similar distinction
regarding love, even though love isn’t something we normally think of as
involving motivation in the way that action does. In light of this, I’ll
henceforth use the term personal reasons rather than the more action-­
specific ‘motivating reasons’.
This also raises a question, though. If it isn’t just their connection with
motivation that makes personal reasons distinctive, what is it? A reasonable
first pass, I think, is that personal reasons explanations provide a distinc-
tively interpersonal form of understanding. What explanations in terms of
a person’s own reasons distinctively do is to enable us to understand things
like actions, thoughts, and emotions from the point of view of their agent
or subject. In particular, they enable us to see the actions, thoughts, and
feelings that they explain not merely as things that were, say, likely to hap-
pen given the conditions mentioned in the explanation, but rather as
things that at least to some extent made sense to the person whose actions,
thoughts, or feelings they were. For now, we can think of personal reasons
as serving a specific interest, an interest in interpersonal, ‘empathetic’
understanding. We will look at one way of explaining what’s distinctive
about this kind of understanding, and hence about personal reasons, in
the next section. First, though, we need to consider the final term of our
tripartite distinction: justifying reasons.
Consider again my action of breaking the cup. We’ve so far considered
a broader and a narrower sense of the question, ‘Why did you do that?’
Another kind of question you might reasonably ask concerns what phi-
losophers call the normative status of my action: whether it was appropri-
ate, sensible, worthwhile, right, justified, and so on, or, put in the language
of reasons, whether there was good reason for me to do what I did. Our
interest in asking such questions is not primarily a concern with why the
action was taken, but with whether the action should have been taken, or
whether there was anything to be said in favor of taking it.
There will often be a certain overlap between justifying reasons and
personal reasons. If I believe something on the basis of a sound inference,
then the premises of that inference should both explain and justify my
belief. Nonetheless, the interests of justification and interpersonal

common than we would like to admit, it probably shouldn’t be our paradigm of believing for
a reason.
182 E. PHILLIPS

understanding are in principle distinct, as can be illustrated by the case of

the cup. The explanation of my breaking the cup in terms of your having
slighted me makes my action perfectly intelligible: you can understand
what point I saw in doing what I did. Yet, accepting the explanation as an
explanation, you might still object that I shouldn’t have done what I did.
A perceived insult, you might say, is no good reason to destroy someone
else’s property. My action was not justified. If, on the other hand, I
knocked the cup off the table in order to distract the assassin who was
sneaking up behind you, this might justify my action—to make my taking
that course of action right or appropriate.
Again, the same distinction also applies to attitudes such as belief and
emotion. The emotions are of particular relevance for our purposes in this
chapter: love, while perhaps not an emotion in itself, is clearly a passion in
a way that beliefs and actions are not. Take again my fear of crossing the
road. It’s plausible that fear is in general appropriate or fitting when the
object of fear is threatening, dangerous, or fearsome (Deonna and Teroni
2012). So there is a kind of normative standard that applies to fear. We can
think of justifying reasons for fear as considerations that bear on whether
this standard is met in a given case. In our example, it is indeed dangerous
to cross the road because of the density and speed of the traffic: the density
and speed of the traffic are reasons for me to feel afraid. By contrast, if I’m
afraid of a perfectly harmless spider, there is no such justifying reason: my
fear might be perfectly intelligible, but it isn’t fitting or appropriate.
Having distinguished these different interests at play in reasons-talk—
explanation, first-personal intelligibility, justification—can we start to
make better sense of the difficulty we faced in accounting for reasons for
love? One obvious suggestion is that different accounts of reasons for
love—as personal qualities, as historical relationships, as mere humanity—
emphasize one or other of the different interests of reasons-talk.
To take a specific example, the central challenge to the simple quality
view—the main proponents of which present it primarily in terms of one
person’s reasons for loving another, or the reasons for which one person
loves another (Delaney 1996; Keller 2000), suggesting that it is intended
in the first instance as an account of personal reasons—was that it fails to
explain why love does not accept substitutes, because if qualities are your
reason for loving one person, then they should equally count as reasons
for loving another person. Note that this move only makes sense if we are
talking about reasons to love, that is to say justifying reasons. If the prop-
erties of a particular loaf of bread at the bakery make it suitable to buy,
 9 SENTIMENTAL REASONS 183

then another qualitatively identical loaf of bread will be just as suitable to

buy. So, all else being equal, rationally speaking, I ought to be indifferent
between the two loaves. If I in fact buy the first loaf, though, then its
qualities may be among my personal reasons for buying it, but I have no
personal reasons for buying the second loaf (at least not in the sense of
‘personal reasons’ sketched earlier), simply because I didn’t buy it: there is
no action of buying the second loaf to be explained. If all that we wanted
the qualities of the beloved to do, then, in their capacity as ‘reasons for
love’, was to provide a special kind of understanding of love—understand-
ing love from the lover’s point of view—it’s not at all clear that the non-
substitutability problem would be any problem at all.
Perhaps, then, the resolution to our puzzle about reasons for love is
simply to treat the distinct interests of justification, explanation, and
understanding separately. This would, however, involve a significant
departure from the way that many have proposed to understand reasons.
While we can in principle distinguish justifying, explanatory, and motivat-
ing reasons for action, belief, and emotion, it is nonetheless attractive to
think that there is also a certain unity or interdependence between them.
One way to capture this unity is in the idea that action, belief, and emotion
are responsive to reasons.

9.4   A Confluence of Interests: ‘Responding’

to Reasons

Personal and justifying reasons, as we have already observed, can overlap.

Perhaps this is more than a mere coincidence. In cases where a reason that
justifies an action, belief, or emotion also makes it interpersonally intelli-
gible, it often seems that the reason serves the latter role partly in virtue of
serving the former: sometimes an action (for instance) ‘makes sense’ to a
person precisely because it is something that that person has good reason
to do. Specifically, this dovetailing of interests seems to occur when some-
one does something for a good reason, or because there is good reason to do
it. You believe that it has rained because the streets are wet: you believe
that it has rained because you recognize the wet streets as good evidence
that it has rained. The wetness of the streets not only both justifies your
belief and explains it from your point of view, but, it seems plausible to say,
it does the latter in virtue of doing the former. This kind of connection
between justifying and personal reasons suggests another way of thinking
184 E. PHILLIPS

of what’s distinctive about the latter. Actions, beliefs, and emotions are all
subject to justifying reasons: each action, belief, or emotion has consider-
ations that count for or against it. It’s natural to think that ideally, our
doing, thinking, and feeling should be appropriately sensitive to such con-
siderations. We should, ideally, do these things on the basis of good rea-
sons. Our actions and attitudes, that is, should be responsive to their
justifying reasons. To respond to a justifying reason in this way, you first
need to be aware of it. If, as many philosophers have argued, justifying
reasons are facts (see, e.g., Alvarez 2010; Kolodny 2005; Lord 2018;
Parfit 2011; Raz 1986, 1999; Scanlon 1998), then being aware of a justi-
fying reason plausibly means, at the very least, believing that the fact
obtains.6
Crucially, we can respond to our beliefs as reasons even when what we
believe is not true. Where what we believe is not true, though, it cannot
(at least on the view we are here supposing) be a justifying reason. Yet
when we act or think or feel on false beliefs in this way, our actions and
attitudes are still potentially susceptible to the kind of interpersonal under-
standing that I have suggested is provided by personal reasons. Moreover,
even when our beliefs are true, the facts on which we act are sometimes
our personal reasons without being genuine justifying reasons—as, for
example, when I break your precious cup in retaliation to a mild insult.
Personal reasons, then, cannot simply be defined as those justifying rea-
sons in response to which a person does what they do. However, there is
still some hope for understanding what makes personal reasons special by
appeal to the notion of justifying reasons. In particular, we might seek to
understand personal reasons as things that seem to the person like (justify-
ing) reasons for acting, thinking, or feeling as they do in response (Alvarez
2010; Scanlon 1998). It may well be that a minor slight is no reason to
damage someone’s property. It might even be that you didn’t actually
slight me at all and I simply misinterpreted your remark. There was no
justifying reason, then, for me to break your cup. Perhaps, though, I did
what I did because it seemed to me as if there was such a reason: I believed
that you slighted me, and this seemed to me, in my anger, like a reason to
destroy something you value. The special way in which my personal rea-
sons explain my breaking the cup, on this way of thinking, is that they

6
A number of philosophers have argued that it means actually knowing the fact in question
(e.g. Hornsby 2008; Hyman 1999; McDowell 2013; Williamson 2000). Whether this is
right isn’t important for our purposes here, so I will simply focus on belief.
 9 SENTIMENTAL REASONS 185

reveal how there seemed to me, at the time, to be some justification for
acting in that way. This, just maybe, gives us a way of explaining what is
distinctive about personal reasons: they explain a person’s actions or atti-
tudes by showing us how, from that person’s point of view, there was some-
thing to be said for acting, thinking, or feeling as they did (compare
Davidson 1980: 9).
Note how this picture unifies our understanding of justifying and per-
sonal reasons. Our personal reasons are, loosely speaking, just what seem
to us to be justifying reasons. As long as we have a reasonable understand-
ing of what sorts of facts in general count in favor of what kinds of
responses, then it will make sense, given this ‘reasons-responsiveness’ pic-
ture, to address questions about justifying and personal reasons together.
Determining what sorts of things justify responses will tell us a lot about
the reasons for which people respond, and looking at people’s personal
reasons might help us think about what kinds of considerations justify. Of
course, people can make mistakes, often very serious ones, about what’s
justifiable, so we will have to be careful. The point is just that, on this
‘reasons-responsiveness’ picture, the two kinds of reasons are not wholly
separate.
Through the lens of ‘reasons-responsiveness’, we can begin to see why
separating out the different interests of reasons-talk, as I suggested in the
last section, could seem misguided. If love is reasons-responsive in the
present sense, the questions of what justifies love and of what explains it
from the lover’s point of view are not so separate. In saying what justifies
love, we would say a lot about what makes sense of it for the lover; under-
standing how love makes sense to the lover could tell us a lot about what
seems to us to justify love. The difficulties we came across in our overview
of the different theories of reasons for love would not, in this case, be eas-
ily resolved by distinguishing justification and interpersonal understand-
ing, because these two interests, while distinct, are nonetheless intimately
interconnected.
Indeed, the assumption that love is reasons-responsive in something
like the present sense seems implicit in many of the authors we have dis-
cussed. Niko Kolodny comes close to making it explicit, basing some of
his arguments on the premise that what normally sustains emotional con-
cern (which he takes to be partly constitutive of love) is a good guide to
the normative (that is, justifying) reasons for it (Kolodny 2003: 162).
Even where there is no such explicit methodological claim, it is not
uncommon to frame discussion in terms of love’s ‘reasons-responsiveness’
186 E. PHILLIPS

(see especially Abramson and Leite 2011, 2018),7 and more generally to
move somewhat freely between claims about the psychology and intelligi-
bility of love on the one hand and claims about its appropriateness or
justification on the other.
Recall the objections to the quality view of reasons for love: the prob-
lems of Universality, Promiscuity, Trading Up, Inconstancy, and
Nonsubstitutability. Each appeals to what seems like an intuitive judgment
about the psychological character of love. We don’t just think that it’s
wrong, irrational, or inappropriate for love to be transferred, to treat one’s
beloved as fungible, or to be ready to ‘trade up’; rather, we think (and this
is reflected in the way that the challenges are typically expressed) that an
attitude that behaved in this way would not be love. This is a claim about
the nature of love, about what it is to love someone. Yet the objections
presuppose that this must be explained in terms of the kinds of justifying
reasons to which love is subject. Similarly, one major problem with the
idea that someone’s mere humanity or personhood is sufficient reason to
love them was that it doesn’t provide an explanation (first- or third-­
personal) of why we love the particular people we love—it doesn’t give us
a satisfactory account of personal reasons for love. There seems, then, to
be an underlying assumption to the effect that the way love behaves psy-
chologically must be explained in relation to the justifying reasons to
which it is sensitive. An obvious explanation for that assumption is the
further assumption that in so far as we love for reasons, love must be
reasons-­responsive in something like the sense I have sketched in this
section.

9.5   Love as a Sentiment

But is it plausible that love is reasons-responsive in this sense? Is it plausi-
ble to think of love as sensitive to facts in the same sort of way that actions
and beliefs are? There are, after all, very significant differences between
love and these other ‘reasons-responsive’ phenomena. Consider some of
the characteristic ways in which intentions and beliefs are formed and

7
Abramson and Leite (2018) express skepticism about the analysis of reasons-­responsiveness
as necessarily running via the agent’s judgments or beliefs. Nonetheless, they maintain that
love is reasons-responsive and seem clearly to think that an account of love’s justifying rea-
sons needs to make sense of its psychology in something like the way outlined in the previous
paragraph.
 9 SENTIMENTAL REASONS 187

revised. Through reasoning, deliberation, judgment, and choice, we have

the capacity to consider what reasons there are for and against believing or
doing this or that and then determining what to think or do in light of
these considerations. When relevant new information arises, we can change
our minds, stop doing what we are doing, revise our plans, or change our
beliefs. In part because of this, intentions and beliefs are ‘fragile’ (as
Wollheim 2003 puts it), readily changing in response to changes in our
situation. Part of the reason it makes sense to expect reasons for acting and
believing to play a privileged role in explaining what we think and do is
that we possess such rational ‘control’ over them: generally (and idealizing
somewhat) we wouldn’t be doing what we are doing, or thinking what we
are thinking, if there didn’t seem to us to be good reason for so doing or
thinking.
Love, at the very least, looks considerably less ‘fragile’—more substan-
tial, we might say—than intentions and beliefs. Love seems to belong to a
category of psychological states sometimes called ‘sentiments’: deep, long-­
lasting, affective orientations toward specific things, which manifest in a
diverse range of ways in our emotions, thoughts, actions, and motivations
(Deonna and Teroni 2012: 106–109; Naar 2017a, 2018). Sentiments are
not formed by choice or judgment but develop gradually over time, typi-
cally through repeated interaction with the thing or person toward which
they are directed. The point here isn’t just that a sentiment takes some
time to ‘set in’ and that once it has done so it tends to stick around, as a
bath might take a long time to fill and a similarly long time to drain.
Rather, it’s that any given instance of a sentiment—one person’s love for
another, a person’s appreciation of a certain genre of music, another’s
distrust of authority—is naturally understood as being shaped by a kind of
developmental history, in something like the way that an individual person
is shaped by their developmental history (compare Rorty 1987; Grau
2010). Indeed, we might say that a good deal of a person’s being shaped
by their history consists in the shaping of their sentiments.
One’s love for another person, in this picture, grows, persists, and
develops through the course of one’s ongoing interaction with that per-
son, one’s thinking about them, and so on. Love can strengthen as it
matures, or it can fade away. An excited romantic love can mature into
something deeper but less emotionally intense; romantic love can change
into platonic affection; a platonic love between friends can turn into some-
thing more romantic. Moreover, if love is a state of this kind, it will natu-
rally tend to persist. It needn’t persist indefinitely, of course—people do
188 E. PHILLIPS

fall out of love—but it seems doubtful that love needs to be sustained by

anything like the lover’s beliefs about why they love the person they love,
in the way that we might think a person’s intention to do something must
be sustained by beliefs about why they are doing what they are doing. If
this is right, it makes a difference to how one person’s love for another will
ordinarily be explained.
Consider, by way of analogy, traits of character or personality. We tend
to think of character as being relatively (not to say completely) fixed, and
hence much less ‘fragile’ than attitudes or stances like beliefs or intentions.
By the same token, a person’s character is not very ‘responsive’ to such
factors as what they believe—including what they believe about how their
character ought ideally to be. If you are a very agreeable person, for
instance, you will not easily become significantly less agreeable, even if you
think it would be better if you were, say because you think you’re too
much of a pushover. This relative insensitivity of character to beliefs is
reflected in how we tend to think a person’s character should be explained.
We have a general picture of what kinds of factors are responsible for char-
acter in which a person’s genetic predispositions, their environment, and
their experiences—particularly their upbringing and experiences in child-
hood—as well as the complex interactions between these factors, play
major roles. Our picture is one on which character is strongly shaped by
events in the individual’s past and by aspects of their biological constitu-
tion much more than by their current beliefs. In so far as character is
indeed shaped by events in a person’s past in this way, character will be
explained developmentally: that is, to make sense of what a person is like
now, we need to take a historical view and try to see how they came to be
that way.
If love is less fragile than belief and intention, if it is the kind of state
that develops over time, we might expect it to be rather more like charac-
ter in this regard too. If so, we should expect love to be less sensitive to
presently obtaining facts that might bear on whether it is ‘appropriate’ or
‘fitting’, or to the lover’s present beliefs about such facts. This, in turn,
suggests that such facts and beliefs will be less privileged in our explana-
tions of love than they are with respect to action, belief, and emotions. It
also suggests that love will be more naturally explained in a historical or
developmental way than are things like belief and intention. While this
distinction is probably best viewed as one of degree rather than an abso-
lute categorical difference, it is nonetheless significant for our thinking
about love and reasons.
 9 SENTIMENTAL REASONS 189

The distinction between these different kinds of explanation can be

nicely illustrated by considering an account of reasons for love that might
superficially seem to be supported by the idea that love exhibits historical
development: Kolodny’s relationship view. Historical relationships play a
central role in Kolodny’s account, but still love is not, in the picture he
presents, explained historically. The reason one has for loving another per-
son, on Kolodny’s account, is simply the fact that one has a certain kind of
relationship with that person. Love is causally sustained by one’s ongoing
recognition of that fact. The history itself explains love at best indirectly,
through the lover’s belief that there is such a history. The changing char-
acter of love over time is explained by the lover’s changing beliefs about
the nature of their relationship with their beloved, not directly by the
events in which the relationship consists. This contrasts with the idea that
events themselves, especially emotionally powerful experiences, leave their
marks on us and on our emotional and affective dispositions—marks that
do not depend wholly on the sustained influence of beliefs about the
events in question.8
The more we see love as explained developmentally, the less it seems to
fit the model of reasons-responsiveness outlined earlier: if history is
explanatorily significant in this way, the lover’s present beliefs have only
limited influence on it. This could be taken to favor a ‘no reasons’ view of
love. If we assume that only reasons-responsive attitudes are subject to
reasons and, further, that something like our model of reasons-­
responsiveness is correct, this would imply that there are no reasons for
love. Indeed, a recent paper by Yongming Han argues against the view
that there are justifying reasons (of the kind that he calls ‘fittingness rea-
sons’) for love precisely on the basis that such reasons plausibly do not play
an important role in explaining why we love the people we love (Han
forthcoming; see also Smuts 2014). Similarly, if reasons-responsiveness is
the only way to make sense of the distinction between personal and merely
explanatory reasons, then if we doubt love’s responsiveness to reasons, we
will thereby doubt the existence of personal reasons for love.
One way to try to avoid these skeptical conclusions would, of course,
be to insist that love is reasons-responsive, perhaps by seeking to articulate
a less demanding account of reasons-responsiveness. I won’t try to assess
the prospects for that line of reply here. What I instead want to do, in the
final section, is to suggest an alternative way forward.

8
Contrast, with Kolodny’s account, the resolutely historical view of love in, for instance,
Grau (2010).
190 E. PHILLIPS

9.6   Divergent Interests

We’ve seen that there are serious difficulties for the idea that love is
reasons-­ responsive in the sense outlined in Sect. “A Confluence of
Interests: ‘Responding’ to Reasons”. Yet something in the claim that there
are no reasons for love rings false. A straightforward explanation of this is
that, even if they do not align in the way that they do in the ‘reasons-­
responsiveness’ model, nonetheless the characteristic interests expressed in
our talk of reasons—in particular, the interest in interpersonal understand-
ing—do apply to love. Love still, it seems, characteristically ‘makes sense’
for the lover in a distinctively first-personal way, and we might still seek to
understand another person’s love from that first-personal perspective. And
it is at least not obvious that love’s non-responsiveness to justifying rea-
sons would mean that there was no sense in which love can be more or less
appropriate or fitting. However, if love’s responsiveness is indeed limited,
if putative justifying reasons do not tend to play the major role in explain-
ing love, then it may be that in order to do justice to the interests expressed
in the language of ‘reasons’—to understand what makes love appropriate
or inappropriate; to make sense of love from the lover’s perspective—we
cannot expect them to fall so neatly together.
Adopting this approach would mean refraining from the assumption
that there is a unified set of concerns to be addressed through an account
of ‘the reasons for love’. Instead, we would recognize that there are ques-
tions about justification and appropriateness on the one hand and ques-
tions about psychology, phenomenology, and understanding on the other.
There might, of course, be connections between these concerns, but the
nature of those connections will itself be something that needs to be
worked out. Crucially, we would not assume that the interesting psycho-
logical aspects of love are necessarily secondary to, or derivative from, facts
about what justifies it. The possibility of separating the respective interests
of justification and understanding in this way suggests a reexamination of
some of the accounts of reasons for love we looked at at the start of this
chapter. For instance, one objection we raised against an account like
Setiya’s, on which love is justified simply by the other’s mere humanity,
was that it fails to give a satisfactory picture of why we love the particular
people we love. Perhaps it is simply a mistake to expect that from an
account of what justifies love. Objections to an account like Setiya’s would,
on this approach, need to take a more directly normative form: for instance,
showing that there are some people whom it would be inappropriate
to love.
 9 SENTIMENTAL REASONS 191

This approach would also suggest that ‘What are the reasons for love?’
is unlikely to be the most perspicuous framing for the interesting issues
concerning our first-personal understanding of love. Once we drop the
assumption that love must be made intelligible by apparent justifying rea-
sons, there is no obvious reason to expect that there is one single kind of
consideration that makes love intelligible.9 The pressing challenge will be
to explain what the ‘intelligibility’ of love consists in if not the kind of
subjective justification that, on the reasons-responsiveness picture, consti-
tutes the first-personal intelligibility of intentional action and belief. Note,
though, that there is good reason to think that our first-personal perspec-
tive on love will be of a different character from that which we have on our
beliefs and intentional actions. In as much as love is characteristically
explained in historical terms, the facts that explain one’s love for another
may be beyond one’s ken in a way that the ‘apparent reasons’ that charac-
teristically explain actions and beliefs are not. In as much as we understand
our love for others historically, then, our self-understanding may in this
regard involve a much greater degree of speculation, vagueness, and
storytelling.
However, there is another, perhaps deeper, thought about intelligibility
and understanding that we might pursue once we depart from the assump-
tion that love must be made intelligible by (apparent) justifying reasons to
which it is a response: namely, that making one’s love intelligible to
another may not be primarily a matter of giving an explanation of why one
loves—in other words, that personal reasons may not after all be a subset
of explanatory reasons. If, for instance, we think of love as involving a kind
of appreciation of the beloved, as some authors have suggested, then talk-
ing in the right way about what you appreciate in your beloved might help
another to understand your perspective as lover in the sense that they can
come to see what you see in your beloved (compare de Sousa 2015, chap.
4)—even if such features aren’t major factors in explaining why you love
that particular person.
In recommending this approach to addressing the set of questions rep-
resented by talk of ‘reasons for love’, I don’t mean we should assume that
there will be no interesting connections between our interests in justify-
ing, explaining, and understanding love. The suggestion is rather that the
differences between love on the one hand and action and belief on the
other are great enough and of such a kind that we should not, at least in

9
Compare Fogal’s (2018) ‘deflationary pluralism’ about motivating reasons for action.
192 E. PHILLIPS

the first instance, assume that the best way to understand love is to try to
fit it into a picture of reasons devised primarily to characterize action and
belief. Indeed, given the extent to which the language of ‘reasons’ has
come to evoke that kind of picture, a final methodological suggestion
might be that this language is, as far as possible, better avoided. Even if we
don’t endorse a ‘no reasons’ view of love, we might consider trying out a
‘no “reasons”’ methodology in thinking about it.10

References
Abramson, Kate, and Adam Leite. 2011. Love as a Reactive Emotion. The
Philosophical Quarterly 61 (245): 673–699. https://doi.
org/10.1111/j.1467-­9213.2011.716.x.
———. 2018. ‘Love, Value, and Reasons’. In The Oxford Handbook of Philosophy
of Love, ed. by Christopher Grau and Aaron Smuts. Oxford: University Press.
https://doi.org/10.1093/oxfordhb/9780199395729.013.7
Alvarez, Maria. 2010. Kinds of Reasons. Oxford: Oxford University Press.
Chang, Ruth. 2011. Can Desires Provide Reasons for Action? In Reason and
Value: Themes From the Moral Philosophy of Joseph Raz, ed. R. Jay Wallace, Philip
Pettit, Samuel Scheffler, and Michael Smith, 56–90. Oxford: Clarendon Press.
Davidson, Donald. 1980. Essays on Actions and Events. Oxford: Oxford
University Press.
De Sousa, Ronald. 1987. The Rationality of Emotion. London: MIT Press.
———. 2015. Love: A Very Short INTRODUCTION. Oxford: Oxford
University Press.
Delaney, Neil. 1996. Romantic Love and Loving Commitment: Articulating a
Modern Ideal. American Philosophical Quarterly 33 (4): 339–356.
Deonna, Julien A., and Fabrice Teroni. 2012. The Emotions: A Philosophical
Introduction. Abingdon: Routledge.

10
I started work on this chapter while a postdoc at the University of Fribourg, working on
the Swiss National Science Foundation-funded project Modes and Contents and as a mem-
ber of the Thumos research group in the Swiss Center for Affective Sciences, University of
Geneva. I would especially like to thank Fabrice Teroni for encouraging me to think about
some of the issues in this chapter, as well as for encouraging me so much in general. An early
version was presented at the Slippery Slope Normativity Summit in Lillehammer, and I
would also like to thank the audience at that conference for a number of insightful questions
and suggestions. Finally, I am grateful to Simon Cushing for very helpful written comments
that greatly improved the quality of the chapter.
 9 SENTIMENTAL REASONS 193

Fogal, Daniel. 2018. Deflationary Pluralism About Motivating Reasons. In The

Factive Turn in Epistemology, ed. Veli Mitova. Cambridge: Cambridge
University Press.
Grau, Christopher. 2010. Love and History. The Southern Journal of Philosophy 48
(3): 246–271. https://doi.org/10.1111/j.2041-­6962.2010.00030.x.
Han, Yongming. forthcoming. Do We Love for Reasons? Philosophy and
Phenomenological Research. https://doi.org/10.1111/phpr.12638.
Hornsby, Jennifer. 2008. A Disjunctive Conception of Acting for Reasons. In
Disjunctivism: Perception, Action, Knowledge, ed. Adrian Haddock and Fiona
Macpherson, 244–261. Oxford: Oxford University Press.
Hyman, John. 1999. How Knowledge Works. Philosophical Quarterly 50
(197): 433–451.
Jollimore, Troy. 2011. Love’s Vision. Princeton: Princeton University Press.
Keller, Simon. 2000. How Do I Love Thee? Let Me Count the Properties.
American Philosophical Quarterly 37 (2): 163–173.
Kolodny, Niko. 2003. Love as Valuing a Relationship. The Philosophical Review
112 (2): 135–189.
———. 2005. Why Be Rational? Mind 114 (455): 509–563. https://doi.
org/10.1093/mind/fzi509.
Korsgaard, Christine M. 1996. The Sources of Normativity. Cambridge: Cambridge
University Press.
Kraut, Robert. 1987. Love De Re. Midwest Studies in Philosophy 10 (1): 413–430.
https://doi.org/10.1111/j.1475-­4975.1987.tb00549.x.
Lord, Errol. 2018. The Importance of Being Rational. Oxford: Oxford
University Press.
McDowell, John. 2013. Acting in the Light of a Fact. In Thinking About Reasons:
Themes from the Philosophy of Jonathan Dancy, ed. David Bakhurst, Margaret
Olivia Little, and Brad Hooker, 13–28. Oxford: Oxford University Press.
Na’aman, Oded. 2015. Reasons of Love: A Case Against Universalism About
Practical Reason. Proceedings of the Aristotelian Society 115 (3pt3): 315–322.
Naar, Hichem. 2017a. Sentiments. In The Ontology of Emotions, ed. Hichem Naar
and Fabrice Teroni, 149–168. Cambridge: Cambridge University Press.
———. 2017b. Subject-Relative Reasons for Love. Ratio 30 (2): 197–214.
https://doi.org/10.1111/rati.12128.
———. 2018. Love as a Disposition. In The Oxford Handbook of Philosophy of Love,
ed. Christopher Grau and Aaron Smuts. Oxford: Oxford University Press.
———. forthcoming. The Possibility of Fitting Love: Irreplaceability and
Selectivity. Synthese. https://doi.org/10.1007/s11229-­018-­02079-­4.
Parfit, Derek. 2011. On What Matters. Vol. 1. Oxford: Oxford University Press.
Protasi, Sara. 2016. Loving People for Who They Are (Even when They Don’t
Love You Back). European Journal of Philosophy 24 (1): 214–234. https://doi.
org/10.1111/ejop.12077.
194 E. PHILLIPS

Raz, Joseph. 1986. The Morality of Freedom. Oxford: Clarendon Press.

———. 1999. Engaging Reason: On the Theory of Value and Action. Oxford:
Oxford University Press.
Rorty, Amelie Oksenberg. 1987. The Historicity of Psychological Attitudes: Love
Is Not Love Which Alters Not When It Alteration Finds. Midwest Studies In
Philosophy 10 (1): 399–412. https://doi.org/10.1111/j.1475-­4975.1987.
tb00548.x.
Scanlon, T.M. 1998. What We Owe to Each Other. London: Harvard University Press.
Schroeder, Mark. 2007. Slaves of the Passions. Oxford: Oxford University Press.
Setiya, Kieran. 2014. Love and the Value of a Life. The Philosophical Review 123
(3): 251–280. https://doi.org/10.1215/00318108-­2683522.
Smuts, Aaron. 2014. Normative Reasons for Love, Part II. Philosophy Compass 9
(8): 518–526.
Velleman, J. David. 1999. Love as a Moral Emotion. Ethics 109 (2): 338–374.
Williamson, Timothy. 2000. Knowledge and Its Limits. Oxford: Oxford
University Press.
Wollheim, Richard. 2003. On the Freudian Unconscious. Proceedings of the
American Philosophical Association 77 (2): 23–35.
Zangwill, Nick. 2013. Love: Gloriously Amoral and Arational. Philosophical
Explorations 16 (3): 298–314. https://doi.org/10.1080/13869795.2013.
812739.
 CHAPTER 10

Wouldn’t It Be Nice: Enticing Reasons

for Love

N. L. Engel-Hawbecker

10.1   Introduction
People ordinarily accept what we can call a Reasons view of love: they
think we can have more or less good reasons to love one another, and
these reasons can move us to do so.1 When we wonder why a friend has
fallen in love with someone who is, to us, a repulsive boor, we look for
potential reasons she has to do so—considerations that bring out this per-
son’s value and might motivate someone of a healthy mind to love him.
Learning of his hidden sweet character, or of the two’s long history
together, we begin to understand the emotional investment in this per-
son’s well-being, the vulnerability to his reciprocal regard, and all the
other elements of our friend’s love. In this way, we can understand some-
one’s love (be it for a romantic partner, a friend, or a family member)

1
For defenses of the Reasons view, see Adams (1999), Keller (2000), Solomon (2001),
Kolodny (2003), Velleman (2006), Abramson and Leite (2011, 2017), Jollimore (2011,
2017a, b), Setiya (2014), Brogaard (2015), Naar (2015), Protasi (2016), Hurka (2017),
Clausen (2019), and Kroeker (2019).

N. L. Engel-Hawbecker (*)
Department of Philosophy, The University of Texas at Austin, Austin, TX, USA

© The Author(s), under exclusive license to Springer Nature 195

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_10
196 N. L. ENGEL-HAWBECKER

sympathetically, as we do their beliefs or intentions—by seeing it as a

response to reasons for it.
This Reasons view of love can seem like common sense at first, but on
reflection it can also seem over-intellectualized, even perverse. For exam-
ple, we ordinarily think that anyone who shares our evidence should be
moved to share our beliefs. But if our friend shares with us her reasons for
loving the boor, she needn’t jealously suppose that we should develop
feelings for him too. For another matter, our friend knows and will con-
tinue to meet people who are at least as charming as this boor—people she
would then seem to have as good (if not better) reasons to love too or
instead. But our friend is not required to reapportion her love whenever
someone better walks in the door in the same way that we are required to
adjust our beliefs whenever better evidence comes in. This can all seem
hard to square with the Reasons view. After all, in each case the reasons for
love are supposedly there. If so, anyone who passes up on them would
seem to be making a mistake.
Reflections like these have encouraged some people to adopt a “No
Reasons view” of love, according to which there can be no reasons for
love. We can understand this position as being motivated by a simple two-­
premise argument. The first premise—call it “Requiring Reasons”—is that
reasons can require people to do what they favor. This idea is familiar
enough: whenever the reasons to do something are strong enough, they
make it a mistake to do otherwise. The second premise—call it “Love’s
Prerogative”—is that nothing can require us to love someone. We are
perhaps required to treat certain people lovingly. But if we fail to regard
them with love, this is (however unfortunate) no more a mistake than
indigestion.2
I will argue that the Reasons view should accept Love’s Prerogative but
reject Requiring Reasons. Specifically, I will argue for a version of the
Reasons view according to which reasons for love neither require nor jus-
tify love. But as the No Reasons view gets right, love does not stand in
need of reasons to be permitted; articulations of the Reasons view that
think otherwise thereby reveal themselves to misunderstand why reasons
for love matter to us. No doubt, this all sounds mysterious. It will become
clearer once we see (first) why we should hold onto the Reasons view over

2
This is at least what I take to be the best argument for the No Reasons view. For others,
see Hamlyn (1978), Thomas (1991), Frankfurt (2004), Zangwill (2013), Smuts (2014b;
MS), de Sousa (2015), and Han (2019).
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 197

the No Reasons view, (second) why we should accept Love’s Prerogative

over Requiring Reasons, (third) why a common way of doing this does
not work, and (fourth) why reasons for love motivate us to love. Then we
will be in a position to answer (finally) any confusions and objections.

10.2   What’s So Good About the Reasons View?

The No Reasons view cannot be accused of perversely overintellectualiz-
ing love and its pathologies. Nonetheless, it is itself intolerably counterin-
tuitive. As we already noted, we often enough understand one another’s
love sympathetically, by considering what reasons might motivate that
love. When we understand someone’s love this way, we do so in the same
way that we ordinarily understand their beliefs and actions—not in the
way we understand why their brow sweats or their hair grows. But if the
No Reasons view were correct, this sympathetic understanding would be
illusory. Insofar as we can understand the “reasons for” another’s love, it
is only through detached, mechanistic explanations.
Additionally, we often evaluate particular loves and lovers: if our friend
loves the boor because of his hidden virtuous character, we will consider
her love and herself to be admirable—less so if the love is based merely on
the fun or carnal pleasures the two have together. (This holds for families
too: we are disappointed by children who love their parents for the good-
ies they bring but not for their good character; we are doubly disappointed
when parents love their children in the same way.) So much is familiar
from Aristotle’s Nicomachean Ethics (1155a1–1157b4). It is also some-
what familiar that we want others to have good reasons to love us: despite
the hype surrounding unconditional love, we rarely rest content at that,
and we strive to be more for others than an occasional relief from bore-
dom and loneliness (Solomon 2001: 155–156). Still less often noted is
how much these evaluations matter to us when we look back on our own
love life: if in old age we have found no better reasons for love than we had
for our high-school affections, we must count ourselves either very lucky
or very unfortunate. But, again, if the No Reasons view were correct, all
such evaluations would be baseless. We could compare loves through the
fruits they bear, but we could not consider any to be more well-rooted
than another.
Finally, when we hear of purely biological or neurological explanations
of love that do not cite good reasons for it, we tend to feel that something
essential goes missing. So far as they go, what we have is no more than
198 N. L. ENGEL-HAWBECKER

lust, infatuation, or brute attachment. If the terms of these explanations

were the only ones available to us, then the possibility of love as something
more would seem a mere illusion, disappearing under the microscope.
Admittedly, it is hard to say how love is “more” than any concoction of
lust, infatuation, and attachment. But I take it that love matters to our
lives in a way that these things do not. So the No Reasons view would bar
us not only from sympathizing with and evaluating one another’s love: it
would seem to prevent us from being able to recognize love in any familiar
sense at all. It would leave our world more disenchanted than we have
found it.
Without the Reasons view, then, certain basic and important parts of
our lived experiences would disappear. Perhaps we are doomed to disen-
chantment, but at present, the No Reasons view forces this upon us only
if Requiring Reasons and Love’s Prerogative are both true. We should
therefore see if we have any independent reasons to reject either premise.

10.3   The Requiring Reasons View

Love’s Prerogative might seem to be the weaker link. Given the worries
that motivated the No Reasons view, it can easily seem an overreaction to
say, as this premise does, that nothing ever requires anyone to love. True,
we are not required to love our friend’s beau, and our friend is not required
to love better suitors once she loves the boor. But two measly cases cannot
show that love is never required. Perhaps less drastic explanations are avail-
able for why love is not required in these cases, leaving open that in others
it is a mistake to not love.
For instance, while we are not required to love our friend’s beau, this
might be because a person’s charms give one a reason to love them only if
one is personally acquainted or in a prolonged relationship with them
(Velleman 2006: 107; Naar 2015; Protasi 2016; Setiya 2021). So in shar-
ing her reasons for loving the boor, our friend need not think she is giving
us any reason to love them too. Reasons for love might be, as philosophers
say, “agent-relative.” Similarly, while our friend is not required to change
whom she loves, “trading up” whenever someone more charming walks in
the door, this might be because her reasons are what philosophers call
“exclusionary” or “silencing”: they not only motivate her current love but
also take any reasons she has to love others off the table (Jollimore 2011:
82–93 and 2017a: 9–14; cf. Velleman 2006: 98–105).
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 199

This all seems possible. However, such possibilities are not enough to
defuse the challenge posed by the No Reasons view. So long as it is also
possible that there are other reasons for love that are not agent-relative or
exclusionary, it remains possible (given Requiring Reasons) for jealousy or
infidelity to still be required of lovers. And it, in fact, seems unlikely that
once we love someone, all our reasons to love anyone else are decommis-
sioned. When a lover’s fidelity waivers, it is not likely for no good reason
(though of course their reasons might not be good enough). It also seems
unlikely that all reasons for love arise from prolonged engagements, so
that none were already there and only slowly discovered. Even if someone
in a personal relationship has more reasons to love the other party (one
reason perhaps being the relationship itself), this does not show that other
people can have no reason to love them too. Perhaps reasons for love are
not so easy to spot as a person’s good looks. But other plausible reasons
(their charisma, intellect, moral fiber, etc.) are not always so hidden as to
be appreciable only to a privileged, dedicated few.
So to avoid false requirements to love, the mere possibility of agent-­
relative and exclusionary reasons for it is no match for Love’s Prerogative.
It would be a match if (and, I think, only if) we could argue that (a) the
Reasons view of love is true and (b) it is true only if reasons for love are
always agent-relative and exclusionary. But (b) is a mere bluff: the Reasons
view of love might be saved by rejecting Requiring Reasons and accepting
Love’s Prerogative instead.3

10.4   The Warranting Reasons View

To preclude any illicit requirements for jealousy and infidelity, some
defenders of the Reasons view have rallied behind the idea that reasons for
love can warrant, justify, permit, or make it appropriate, but they never
make not loving a mistake. This might be because reasons for love are
always counterbalanced by reasons against it: perhaps the distrustfulness of
humanity or the threat of heartbreak always gives us sufficient reason to

3
I have focused on two ways to save Requiring Reasons, but the problem generalizes. It is
also not enough to say that some reasons for love are personal relationships (Kolodny 2003;
Hurka 2017), incomparable values (Velleman 2006), indeterminate developments (Bagley
2015), or organic unities (Clausen 2019). So long as there might be other kinds of reasons
for love, Requiring Reasons raises the threat of required jealousy and infidelity. And the pos-
sibility of these other reasons cannot be dismissed on the ad hoc grounds that they would cast
doubt on the Reasons view committed to Requiring Reasons.
200 N. L. ENGEL-HAWBECKER

keep to ourselves. Or, more plausibly, the idea might be that reasons for
love are simply not in the business of requiring our love. To require some-
thing is to permit it and forbid its absence. But a consideration that speaks
in favor of one option need not obviously speak against (let alone forbid)
any other. So we should avoid equating the ability to permit a response
with the ability to forbid others.4 We might expect, then, that there are
considerations with the first ability but not the latter—merely warranting
reasons, as we might call them. This naturally leads us to the proposal that
reasons for love are of this warranting rather than the requiring sort.5
This Warranting Reasons view, as we might call it, denies Requiring
Reasons and accepts Love’s Prerogative. If true, it allows us to maintain all
the insights of the Reasons view: we still have reasons for more than lust,
infatuation, or attachment; these reasons redound on the quality of our
love and our character; and we can still sympathetically understand one
another’s love. At the same time, there can be no threat of such reasons
ever requiring us to indulge in jealousy or infidelity. The Warranting
Reasons view, then, answers the No Reasons view’s challenge.
Nevertheless, this approach runs into four problems. The first is that we
cannot infer that reasons for love merely warrant it from the fact that it
would be nice for the Reasons view if they do. That would be wishful
thinking. We again need an independent argument for the claim that all
reasons for love must be warranting rather than requiring. Without such an
argument, the possibilities of perverse requirements remain, and the No
Reasons view will still seem like a necessary revision to common sense.
And at this point, the Reasons view cannot rest content with the sugges-
tion that reasons for love are all either warranting, agent-relative, or exclu-
sionary. Once the only way to save common sense becomes so ad hoc,
revisionism (here, the No Reasons view) becomes credible.
The second problem is that if love needs reasons before it is permitted,
there must be something forbidding it. It cannot be forbidden merely by
the absence of reasons supporting it: we constantly do all sorts of small
things without any reason, yet we do not thereby commit even minor
mistakes. Rather, what forbid us from doing something are only ever
4
For more on this idea, see Gert (2003), Greenspan (2005), Little (2013), Scanlon (2014:
107), Little and Macnamara (2017), Darwall (2017), and Whiting (2020).
5
This proposal is made by Adams (1999: 163), Abramson and Leite (2011, 2017),
Jollimore (2017a), and Brogaard (2015: 78). Kolodny (2003) and Setiya (2014) both con-
sider (with more or less sympathy, respectively) that reasons for love might be like this;
however, they frame this idea in Kagan’s (1989) terminology of “non-insistent reasons.”
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 201

substantive decisive reasons against doing it. So if we say that reasons for
love permit it, we must admit that there are otherwise decisive reasons
against love. But it is not at all clear there ever are any decisive reasons
against love per se (pace Driver 2014). If someone is violent or cruel, that
is a good reason to stay away from them but not a decisive reason to stop
loving them entirely.6 If someone is a total monster, this seems like a rea-
son to treat them in certain ways so as to prevent them from doing evil.7
Their monstrosity might also be a reason to invest our other emotional
resources elsewhere, so that our own well-being is not tethered solely to
such an ill-­fated prospect. But these precautions are compatible with lov-
ing the “unlovable,” albeit from a sensible distance (literally and
figuratively).8 Furthermore, most of us are intimately aware of the reasons
we give others not to love us. Yet we hope that we might be loved despite
such reasons: we may hope for this even when we think we give another
overwhelming reason not to love us. If such love would be wrong on the
other’s part, this hope would be perverse. But it is not: what is hoped for
is something undeserved but not impermissible—a form of grace, akin to
gifted forgiveness and the supererogatory (cf. Darwall 2017: 98). In sum,
if there are warranting reasons for love in some cases, there must also be
some decisive reasons against loving elsewhere. But it is doubtful that the
latter exist.
If nothing forbids love, then this absence permits it—not any reasons
for love. This brings us to the third problem for the Warranting Reasons
view: generally, what permits something and what favors it can be distinct
considerations. Consider, for example, a judge’s search warrant. It permits
the search of a citizen’s property but does not itself give officers a reason
to conduct the search. What favors the search is some probable cause. But
a probable cause does not itself make the search appropriate. For another
6
It is often said that love constitutively involves a wish to be near the beloved. But this
“truth” is likely a defeasible generic, often defeated in ordinary adult relations. As Velleman
(2006: 86) writes, “When divorcing couples tell their children that they still love one another
but cannot live together, they are telling not a white lie but a dark truth.”
7
While it is commonly said that lovers want to promote their beloved’s ends on their
behalf, this seems to be at most a generic truth, which Ebels-Duggan (2008) argues against
at length. So we should not assume that loving a monster requires facilitating their misdeeds.
8
Setiya (2014: 257–258) seems sympathetic to this point. It is also worth bearing in mind
that love is compatible with plenty enough hostility (Neu 2000). Because of this, there is an
obvious difference between a sensible love for someone despite their flaws and a blind love
that ignores or condones such flaws. What is wrong with the latter is not the love but the
blind condoning (for a similar point, see Smuts 2014b: 523).
202 N. L. ENGEL-HAWBECKER

example, Little and Macnamara (2017) liken warranting reasons to tickets

that allow us to board a plane. This analogy is instructively inapt, since
tickets do not give us any reason to board a plane: they merely allow one
to board. What gives one a reason to board is the destination, yet that
does not allow one to board. Just so, what permits love and what counts
in favor of it are two separate things. Love is permitted by the absence of
anything forbidding it. But this absence does not itself count in favor of
love. What does that is some quality of the beloved, their character, one’s
shared history with them, one’s potential future with them, or what-have-­
you. And these reasons for love cannot make love further permitted, since
permission is, strictly speaking, all or nothing (Maguire 2017: 791–2). So,
given that love is already permitted by something that is not itself a reason
for love, we cannot say that love is permitted (warranted, justified) by
reasons for it.

10.5   The Enticing Reasons View

This brings us to the fourth and most important problem with the
Warranting Reasons view: by encouraging us to look for justifications for
love, it makes us lose sight of what reasons for love do to and for us. The
Warranting Reasons view is not alone in this. While everyone agrees that
reasons are considerations that favor certain responses, there is a constant
temptation to say more than this. For example, it is common to hear that
reasons “tell us what we should do, feel or believe” (Kroeker and
Schaubroeck 2017: vii; emphasis added; cf. Smuts 2014a: 507). But we
might recognize that this is too strong: it precludes the possibility of war-
ranting reasons, which would tell us what we may do without suggesting
it would be a mistake to do otherwise. So it might instead be said that
“reasons… count in favor of actions by rationalizing and justifying them”
(Kroeker 2019: 278, emphasis added; cf. Kroeker and Schaubroeck 2017:
viii); we may be told that the “kinds of normative reasons we are after are
those of appropriateness. We want to know if love can be justified” (Smuts
2014a: 507; emphasis added). These remarks leave open the possibility of
merely warranting reasons. But they still clearly assume—as is common—
that reasons can rationalize a response or make it intelligible only if they
also justify it, permit it, or make it appropriate.
This assumption, however, is falsified by the cases we just saw, where a
response is favored and made appropriate (permitted, justified) by two
distinct things. For a new example, take games: what permits a move in
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 203

one will be the rules or a referee, but what favors the move will be some-
thing like the advantages it brings. To rationalize or make sense of a move,
we cite considerations of the latter sort—not the former.
But worse than being false, the assumption that reasons favor and ratio-
nalize responses only if they help justify it is deeply misleading. It encour-
ages us to conflate a reason’s ability to favor a course of action with its
ability to permit it. (This conflation is manifest in the above quote that
says reasons favor actions partly by justifying them.) If considerations of
value (how something or someone is attractive or unattractive, fun or bor-
ing, tasteful or distasteful, etc.) mattered only insofar as they opened up or
closed off certain courses of action, then practical reasoning would seem
to be no more than the boring application of otherwise pointless rules.
(The feeling that it is nothing more than this is likely familiar to anyone
who has encountered ethical theorizing modeled on legal theory.) The
appeal of any given response then goes missing, and so the point of com-
plying with our reasons reduces to either, “It is the only option left avail-
able to me” or, worse, “I had these options open to me, so I had to just
pick one.” This is all to say, if the relevance of reasons were exhausted by
their ability to erect or remove barriers between what is right and wrong,
permissible or not, then their ability to motivate us and make our responses
somewhat sympathetic (intelligible, rationalizable) would disappear from
sight. By taking reasons to be good only for justifications, we lose sight of
why they matter to us and set ourselves up for another sort of
disenchantment.
We earlier distinguished how a consideration can permit a response
from how it can forbid others. These deontic abilities should be distin-
guished from how a consideration can bring out the value of something.
This evaluative ability may operate in tandem with the deontic ones, but
it is distinct. A consideration that has the evaluative ability alone can be
called an enticing reason, which is in the business of merely “making an
option attractive rather than demanded, required, or right” (Dancy
2004: 91).9

9
The language of “enticing reasons” was first introduced by Raz (1999), who doubted
their possibility. Their possibility has also been challenged by Robertson (2008) and Nebel
(2018) but defended by Dancy (2004) and Little (2013). Since this literature concerns the
general question of whether there could be any enticing reasons for anything, it does not
provide much help in specifying what specifically we have enticing reasons for. I have already
deployed one heuristic implicit in Dancy’s work: if some response is already permitted by the
absence of anything forbidding it, then any reasons for it seem to be enticing (unless they are
204 N. L. ENGEL-HAWBECKER

If reasons for love were enticing, this would again allow the Reasons
view to avoid any illicit requirements for infidelity or jealousy. It would
also free the Reasons view of any commitment to find decisive reasons
against love, which would need to be counterbalanced before love is per-
mitted. But, as I have repeatedly insisted, it is one thing to demarcate a
certain class of reasons and quite another to show that all reasons for love
belong to this class. We again need independent evidence showing that all
reasons for love must be merely enticing.
To find such evidence, we might look to why reasons for love motivate
us to love. Presumably, it is no mere coincidence. Rather, these consider-
ations motivate love because they are good reasons for it.10 Does this mean
they motivate love because, inter alia, they help permit it? No: even if (per
impossibile) these considerations did help permit love, that is not why they
motivate it. Permitting love, then, seems to be no part of what it is to be
a reason for love, since it does not help explain why such reasons motivate
love (assuming they do so because of what they are).
Similarly, it does not seem that reasons for love motivate it because they
make not loving a mistake: what does that are reasons against not loving
(considerations of disvalue), which motivate not love but fear of an unlov-
ing life. This fear might dispose us to be more easily moved by our reasons
for love, but it is still then just a background catalyst for love—not the
real thing.
The only obvious option left, then, is that reasons for love motivate it
simply because they bring out the value of the would-be beloved. But this
seems exactly right. It also makes it no coincidence that reasons for love
motivate it: to love someone is to appreciate them in a way, and consider-
ations that bring out their value would be grounds for such
appreciation.11

also reasons against every other alternative, which would seem to make them of the requiring
sort). The argument below provides another test for enticing reasons.
10
For a defense of this explanatory connection when it comes to believing for good rea-
sons, see Lord and Sylvan (2019). Similar proposals are made by Wedgwood (2006) and
Arpaly and Schroeder (2014). If cases of deviant causal motivation seem possible to you,
append “in the right way” each time I use “motivate.”
11
Frankfurt (2004: 39) says he “can declare with unequivocal confidence that I do not love
my children because I am aware of some value that inheres in them independent of my love
for them.” Given the parent–child relation, his love is presumably no mere coincidence. But
we are concerned with when it ceases to be a mere coincidence that certain considerations
motivate love, and Frankfurt is just denying that his love is motivated at all. He is not chal-
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 205

So permitting or requiring love seems to be no part of why reasons for

love motivate it, when they motivate it because of what they are. When we
look at why reasons for love as such motivate it, we see them operating as
enticing reasons. This indicates that reasons for love are essentially entic-
ing, and so it gives us independent evidence for the Enticing Reasons view
of love. Admittedly, this evidence is subtle and delicate, given the perils of
extracting normative conclusions from broadly phenomenological obser-
vations. The point is just that the Enticing Reasons view has something
more to say in its defense than (roughly) it would be nice for the Reasons
view if reasons for love were only ever enticing. This something more pre-
empts any accusation that the Enticing Reason view is just an ad hoc revi-
sion to common sense, no more plausible than the revision offered by the
No Reasons view. And given this evidence that reasons for love are entic-
ing, the No Reasons view cannot fairly press objections that presuppose
reasons for love must warrant or require it.

10.6   The Threat of Overgeneralization

Perhaps the argument just given for the Enticing Reasons view is not so
interesting: it might seem plausible that reasons only ever motivate us
insofar as they present us with something of value (or disvalue). If we can
infer what reasons are like from how they motivate us, then the argument
of the last section would threaten to generalize, showing all reasons to be
merely enticing. And if this result is not false, it at least trivializes an
Enticing Reasons view of love.
To avoid this overgeneralization, it should suffice to find one example
of what might, for all we have said, be a non-enticing reason. For a familiar
contrast with love, consider respect, in the Kantian sense of regarding
another’s humanity as one cannot help but regard one’s own. David
Velleman suggests that “respect and love [are] the required minimum and
optional maximum responses to one and the same value,” this humanity
(2006: 101; cf. Setiya 2014: 262). This suggests that despite stemming
from the same value as love, reasons for respect are of the requiring sort.
If they are, then we would have to say the following: reasons for us to
respect another motivate us to do so because these considerations inter
alia make it a mistake not to do so. This is plausible if one thinks (as

lenging why reasons for love motivate—only that they motivate it. But the Reasons view
anyway says only that reasons for love can motivate it—not that they always do.
206 N. L. ENGEL-HAWBECKER

Velleman does, following Kant) that we have a reason to respect others

because their humanity is no different from our own. Given this, we would
be mistaken to not regard it as we must regard our own: in doing so, we
would be regarding something as unlike itself; at least, we would be
regarding like cases as unalike. Hence, it would be mere bizarre luck if we
were moved to respect others (in the Kantian sense) by good reasons to do
so, but not even partly because these considerations show such respect to
be required. And it is not unfathomable that the threat of error helps
explain why we are moved to respect others in a way that, as Kant puts it,
“does not serve my inclination” (2011: 30–31). Respect can be begrudg-
ing, reluctant, or, as Kant says, akin to fear, which would make little sense
if reasons for respect motivated it simply as considerations of positive value.
This account of respect and its reasons might be wrong, but nothing we
have said rules it out. And since it entails that reasons for respect are not
enticing but requiring, nothing we have said entails that all reasons must
be enticing. (Of course, reasons for much else besides respect may not be
enticing. Reasons for belief are likely not enticing.)

10.7   The Abundance and Inconstancy

of Enticing Reasons

Even if one grants the validity of the argument for the Enticing Reasons
view, one might think this digs the Reason view in only a deeper hole.
After all, the Enticing Reasons view allows that love is always permitted:
rampant infidelity, then, would be allowed. Still worse, many people agree
that we are sometimes required to treat certain people with partiality. A
parent who treats their child no differently than any other obviously does
something wrong. A natural way to explain why this partiality is required
is by holding that it expresses a love that is required. But on the Enticing
Reasons view, love is never required. How, then, could such expressions
be? And if we are free to love and not love anyone, in what sense could we
owe partiality to specific people and not just whomever we happen to love
at the moment? If we need no reasons for love but loving someone can
anyway create duties of partiality, then it seems we can rampantly boot-
strap duties. At the same time, since we never need to love anyone, we
could also illicitly unwind such duties, ceasing to love someone when
times get tough.
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 207

Much of these problems can be handled if we first distinguish between

reasons for love and reasons of love.12 I have argued that the former are
enticing, but I have said nothing about the latter. Reasons of love are
­supposed to include all the reasons that arise in intimate relationships to
treat the other party in special ways; these reasons include all the special
promises, obligations, and expectations we make and undertake in such
relationships. Despite the name, then, it seems likely that reasons of love
come not from love per se but from the relevant histories and relationships
in which love paradigmatically occurs (Velleman 2006: 108–109; Jeske
2017; Sadler 2017). These reasons could be requiring and would be if, as
seems plausible, they reduce to reasons for respect (Smith 2017; Darwall
2017: 99). This is compatible with Love’s Prerogative and the Enticing
Reasons view: to repeat an earlier claim, although we are not required to
regard others with love, we can be required to treat those closest to us
lovingly (cf. Kolodny 2003: n. 6; Driver 2014: 9). But we need not grant
the assumption that such partiality is owed because it is an expression of
owed love.13
Still, even if a person treats everyone as they should, they might seem
to be doing something wrong if their loving regard wavers, say, away from
their spouse and toward their neighbor’s. The Enticing Reasons view can
admit that this would be unfortunate or disappointing. But so long as this
fickle lover in fact behaves themselves, it is not so obvious that they are
doing anything wrong. We feel that they truly err, I think, only if we need-
lessly assume that the burgeoning love for the neighbor’s spouse is roman-
tic. Yet what would be illicit in that case is not the love but the romance—that
is, the relationship with its associated behavior. But if again our fickle lover
behaves themselves, then we cannot assume their new love is that of an
illicit romantic partner rather than that of a mere friend, which would of
course be fine.14

12
For more on the latter, see Pettit (1997), Wallace (2012), and the later essays in Kroeker
and Schaubroeck (2017).
13
Liao (2015) argues that parents cannot adequately discharge their parental duties with-
out actually loving their children (120–123). His arguments presuppose that to treat their
child lovingly enough, unloving parents would need to prevent their child from not just
actually but even potentially discovering their lack of love. Liao is likely right to suggest no
parent could pull this off, but the requirement is anyway unreasonably excessive.
14
Some assume ‘love’ is wholly ambiguous, and romantic, friendly, and familial love are as
distinct emotions as Jack, Bobby, and Ted Kennedy are distinct individuals. I instead assume
‘romantic,’ ‘friendly,’ and ‘familial’ should not be read as the first terms of proper names but
208 N. L. ENGEL-HAWBECKER

This handles one half of the infidelity—the new love for others. But
something must still be said about the other half—the lost love for one’s
partner.

10.8   The Unloving

Since reasons for love never make it a mistake to not love, certain terms of
criticism seem inapplicable to those who recognize but fail to be moved by
them. We cannot call these people mistaken, wrong, unreasonable, irratio-
nal, and so on. Jonathan Dancy (2004) says someone who turns down
their most enticing option is “silly.” This is not likely what we want to call
the unloving. But “silly” seems generally to not be the right label for those
who choose the less enticing option: silliness is just a mild form of irratio-
nality; to be irrational is to be mistaken in a way; and enticing reasons (like
warranting reasons) do not make one mistaken if one turns them down.15
Rather, as I have suggested, a failure to love is unfortunate and disap-
pointing—what Julia Driver (1992) calls “suberogatory.” This does not
show the unloving to be making a mistake, but it does make them appear
cold, dull, obstinate, strange, unfeeling, unsympathetic, or even unfortu-
nate themselves. And these seem to me just the sort of criticisms that are
generally appropriate for those who knowingly turn down their most
enticing option. Perhaps if we were in their position, we expect or at least
hope that we would have taken that option. This explains our disappoint-
ment and why we struggle to understand them or sympathetically imagine
ourselves responding in the same way. But it is noteworthy that we are
now speaking mostly about ourselves. It does not follow that the unloving
are making a mistake. All that follows is that they are unlike us.
While such criticisms seem appropriate for partners who for no reason
cease loving one another, these terms might still seem inapt for other fail-
ures to love. After all, when our friend continues to love her boor but
none of the better suitors who come around, this does not (or anyway

as adjectives locating the relationship in which love resides and manifests—hence my assump-
tion that love outside a romance cannot be romantic. Unfortunately, the temptation to think
otherwise is all the more entrenched given the common conflation of romantic love with lust
or infatuation (which can, of course, occur outside romantic relationships).
15
Perhaps irrationality is not always a mistake. If so, we can call the unloving irrational, but
since this would still be misleading, we would need to elaborate: one will have to appeal to
other cases where irrationality seems alright, as when Huck Finn can’t bring himself to turn
Jim in despite thinking he ought to do so.
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 209

need not) make her appear strange to us. And she will not likely find us
unsympathetic or unfortunate just because we do not love her boor too.
There are two possible (and compatible) explanations for this. First, it
is consistent with the Enticing Reasons view that some reasons for love are
agent-relative and exclusionary. So in the cases above, it might be that
these failures to love are not failures to indulge one’s enticing reasons, as
there are no reasons to indulge.16
This reply, however, is not very interesting: let us set it aside and sup-
pose that in each case, the person who fails to love is genuinely turning
down their most enticing option. For the fact that our friend does not
seem strange to us, or we to her, can be explained away by our expectation
that if we were in each other’s position, we would respond in the same
way. We are like one another, and so we can sympathize with one another’s
responses. This similarity naturally preempts any accusations of being
strange, unsympathetic, cold, and so on.
More precisely, it preempts our accusations. It remains conceivable that
we and our friend will appear alien and cold in the eyes of universal, aga-
peic lovers, who are moved by every reason they find to love those they
encounter. To them, our failure to love everyone we meet would presum-
ably make us appear unfortunate, since we fail to appreciate the value oth-
ers bring to our lives. That perspective, while obviously not our own and
not required (given the Enticing Reasons view), is not unfathomable. This
itself is something any Reasons view should account for, as the Enticing
Reasons view does.

10.9   Conclusion
A theme of this chapter has been that past defenses of the Reasons view
have relied upon a sort of wishful thinking. They provide good arguments
for thinking some (perhaps the best) reasons for love are nuanced (e.g.
warranting, agent-relative, or exclusionary) in ways that would avoid any
false requirements for infidelity or jealousy. But it is hard to find any argu-
ment that all reasons for love are like this, independent of the fact that it
would be nice for the Reasons view if they all were. Presumably, we can
have no credible intuitions about such a general and theoretical claim. Nor

16
The agent-relativity of reasons for love would even seem to be better explained by the
Enticing Reasons view. As Dancy (2004: 107) notes, “What is enticing for one person is not
for another, and properly so. (This is another interesting aspect of the enticing.)”
210 N. L. ENGEL-HAWBECKER

can we credibly claim that whatever lacks these nuances is no reason for
love. But without the universal claim, the No Reasons view’s challenge
stands. Because of this, I have provided an independent, if modest, argu-
ment for the conclusion that reasons for love must be enticing: roughly,
reasons are what reasons qua reasons do, and when reasons for love qua
reasons for love motivate it, they do so qua enticing reasons.
According to this Enticing Reasons view of love, certain considerations
count in favor of love, but that is all they do. They neither justify nor
require love. That is fine, however, since love is already permitted by the
absence of anything forbidding it. To adapt a line of Wittgenstein’s (1953:
Sect. 289), to love without a justification does not mean to love wrong-
fully. In this respect, the Enticing Reasons view agrees with the No Reasons
view. There is something amiss in defending love or feeling the need to.
From this, the No Reasons view seems to conclude that nothing can be
said for love in any particular case. But this inference falsely assumes that
reasons for love must be potential justifications for it. Love does not need
reasons to justify it (as the No Reasons view recognizes), but this does not
mean there cannot still be any considerations favoring and motivating it.
Nor does it follow that these considerations cannot be reasons for love.
They do exactly what we would expect such reasons to do: they help us
sympathetically understand one another’s love as we do their beliefs and
actions; they redound on the quality of love and lovers; and they let us live
in a world where there is more to affection than superficial lust and brute
biological attachments. The No Reasons view might rightly accuse past
articulations of the Reasons view of overintellectualizing love, but all par-
ties to this dispute are guilty of overintellectualizing (or in any case over-
thinking) what reasons for love must be like. They are just considerations
of value, and that is as much as we want.
This brings out a second theme of this chapter, which echoes Plato’s
suggestion (in Book VII of the Republic) that the hardest thing for us to
look at is the Form of the Good: instead of seeing its true nature, we are
more likely to shy away and dwell on mere proxies. It is perhaps not so
surprising that we might similarly shy away from love itself (Cavell 1969).
In any case, I have argued that something like this occurs when we try to
look closely at reasons for love. Instead of staring down considerations of
value, we often divert our attention to permissions instead. This threatens
to make us lose sight of love as we know it. But the problem, then, is not
that we are trying to view love as a response to reasons. The problem is
that instead of maintaining a Reasons view of love, we quickly back down
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 211

and try to see it in view of something else—justifications or requirements.

So this failure does not suggest that a No Reasons view of love is needed
to properly appreciate it; given the No Reasons view, it is not clear what
we would even have to appreciate. Rather, as we began by noting, we get
a view on love by looking at reasons for it.17

References
Abramson, Kate, and Adam Leite. 2011. Love as a Reactive Emotion. The
Philosophical Quarterly 61 (145): 673–699. https://doi.org/10.1111/
j.1467-­9213.2011.716.x.
———. 2017. Love, Value, and Reasons. In The Oxford Handbook of Philosophy of
Love, ed. Christopher Grau and Aaron Smuts. Oxford: Oxford University Press.
https://doi.org/10.1093/oxfordhb/9780199395729.013.7.
Adams, Robert Merrihew. 1999. Finite and Infinite Goods: A Framework for Ethics.
Oxford: Oxford University Press.
Arpaly, Nomy, and Timothy Schroeder. 2014. In Praise of Desire. Oxford: Oxford
University Press.
Bagley, Benjamin. 2015. Loving Someone in Particular. Ethics 125: 477–507.
Brogaard, Berit. 2015. On Romantic Love: Simple Truths about a Complex Emotion.
Oxford: Oxford University Press.
Cavell, Stanley. 1969. The Avoidance of Love: A Reading of King Lear. In Must
We Mean What We Say?: A Book of Essays. Cambridge (UK): Cambridge
University Press.
Clausen, Ginger T. 2019. Love of Whole Persons. The Journal of Ethics 23:
347–367. https://doi.org/10.1007/s10892-­019-­09299-­x.
Dancy, Jonathan. 2004. Enticing Reasons. In Reason and Value: Themes from
Joseph Raz, ed. R. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael
Smith, 91–118. Oxford: Oxford University Press.
Darwall, Stephen. 2017. Love’s Second-Personal Character: Reciprocal Holding,
Beholding, and Upholding. In Love, Reasons and Morality, ed. Esther Engels
Kroeker and Katrien Schau-broeck, 93–109. New York: Routledge.
Driver, Julia. 1992. The suberogatory. Australasian Journal of Philosophy 70
(3): 286–295.
———. 2014. Love and Duty. Philosophic Exchange, 44(1).
Ebels-Duggan, Kyla. 2008. Against Beneficence: A Normative Account of Love.
Ethics 119 (1): 142–170.

17
This chapter has been greatly helped by discussions with Ray Buchanan, Sophie Cote,
Simon Cushing, Bryce Dalbey, Jonathan Dancy, Julia Driver, Matt Evans, Kathleen Higgins,
Emilie Pagano, and David Sosa.
212 N. L. ENGEL-HAWBECKER

Frankfurt, Harry. 2004. The Reasons of Love. Princeton: Princeton University Press.
Gert, Joshua. 2003. Requiring and Justifying: Two Dimensions of Normative
Strength. Erkenntnis 59: 5–36.
Greenspan, Patricia. 2005. Asymmetrical Practical Reasons. In Experience and
Analysis: Proceedings of the 27th International Wittgenstein Symposium, ed.
M.E. Reicher and J.C. Marek, 387–394. Vienna: ÖBV & HPT.
Hamlyn, D.W. 1978. The Phenomena of Love and Hate. Philosophy 53: 5–20.
Han, Yongming. 2019. Do We Love For Reasons? Philosophy and Phenomenological
Research. https://doi.org/10.1111/phpr.12638.
Hurka, Thomas. 2017. Love and Reasons: The Many Relationships. In Love,
Reasons and Morality, ed. Esther Engels Kroeker and Katrien Schaubroeck,
163–180. New York: Routledge.
Jollimore, Troy. 2011. Love’s Vision. Princeton: Princeton University Press.
———. 2017a. Love as ‘Something in Between’. In The Oxford Handbook of
Philosophy of Love, ed. Christopher Grau and Aaron Smuts. Oxford, Oxford
University Press.
———. 2017b. Love: The Vision View. In Love, Reasons and Morality, ed. Esther
Engels Kroeker and Katrien Schaubroeck, 1–19. New York: Routledge.
Kagan, Shelly. 1989. The Limits of Morality. Oxford: Oxford University Press.
Kant, Immanuel. 2011. Groundwork for the Metaphysics of Morals. Translated by
Mary Gregor and Jens Timmerman. Cambridge, UK: Cambridge
University Press.
Keller, Simon. 2000. How Do I Love Thee? Let Me Count the Properties.
American Philosophical Quarterly 37 (2): 163–173.
Kolodny, Niko. 2003. Love as Valuing a Relationship. The Philosophical Review
112 (2): 135–189.
Kroeker, Esther Engels. 2019. Reasons for Love. In The Routledge Handbook to the
Philosophy of Love, ed. Adrienne M. Martin, 277–287. New York: Routledge.
Kroeker, Esther Engels, and Katrien Schaubroeck. 2017. Love, Reasons and
Morality. New York: Routledge.
Liao, S. Matthew. 2015. The Right to Be Loved. Oxford: Oxford University Press.
Little, Margaret Olivia. 2013. In Defense of Non-Deontic Reasons. In Thinking
About Reasons: Themes from the Philosophy of Jonathan Dancy, ed. David
Bakhurst, Brad Hooker, and Margaret Olivia Little, 112–136. Oxford: Oxford
University Press.
Little, Margaret Olivia, and Coleen Macnamara. 2017. For Better or Worse:
Commendatory Reasons and Latitude. In Oxford Studies in Normative Ethics,
ed. Mark Timmons, vol. 7, 138–160. Oxford: Oxford University Press.
Lord, Errol, and Kurt Sylvan. 2019. Prime Time (for the Basing Relation). In In
Well-Founded Belief: New Essays on the Basing Relation, ed. J. Adam Carter and
Patrick Bondy. New York: Routledge.
 10 WOULDN’T IT BE NICE: ENTICING REASONS FOR LOVE 213

Maguire, Barry. 2018. There Are No Reasons for Affective Attitudes. Mind 127
(507): 779–805.
Naar, Hichem. 2015. Subject-Relative Reasons For Love. Ratio 30: 197–214.
https://doi.org/10.1111/rati.12128.
Nebel, Jacob M. 2018. Normative Reasons as Reasons Why We Ought. Mind 128
(510): 459–484. https://doi.org/10.1093/mind/fzy013.
Neu, Jerome. 2000. A Tear is an Intellectual Thing: The Meanings of Emotion.
Oxford: Oxford University Press.
Pettit, Philip. 1997. Love and Its Place in Moral Discourse. In Love Analyzed, ed.
Roger E. Lamb, 153–164. Boulder, CO: Westview Press.
Protasi, Sara. 2016. Loving People for Who They Are (Even When They Don’t
Love You Back). European Journal of Philosophy 24 (1): 214–234. https://doi.
org/10.1111/ejop.12077.
Raz, Joseph. 1999. Explaining Normativity: Reason and the Will. In Engaging
Reason, 90–117. Oxford: Oxford University Press.
Robertson, Simon. 2008. Not So Enticing Reasons. Ethical Theory and Moral
Practice 11: 263–277. https://doi.org/10.1007/s10677-­007-­9091-­5.
Scanlon, T.M. 2014. Being Realistic About Reasons. Oxford: Oxford
University Press.
Setiya, Kieran. 2014. Love and the Value of a Life. Philosophical Review 123
(3): 251–280.
———. 2021. Other People. In The Value of Humanity: A Re-Evaluation, ed.
Sarah Buss and Nandi Theunissen. Oxford: Oxford University Press.
Smith, Michael. 2017. The ‘What’ and ‘Why’ of Love’s Reasons. In Love, Reasons
and Morality, ed. Esther Engels Kroeker and Katrien Schaubroeck, 145–162.
New York: Routledge.
Smuts, Aaron. 2014a. Normative Reasons for Love, Part II. Philosophical Compass
9 (8): 518–526.
———. 2014b. manuscript. In Defense of the No-Reasons View of Love. Obtained
October 2018 from https://philpapers.org/rec/SMUIDO. Accessed
August 17, 2020.
Solomon, Robert. 2001. On Love. Indianapolis: Hackett Publishing Company, Inc.
de Sousa, Ronald. 2015. Love: A Very Short Introduction. Oxford: Oxford
University Press.
Thomas, Laurence. 1991. Reasons for Love. In The Philosophy of (Erotic) Love, ed.
Robert C. Solomon and Kathleen M. Higgins, 467–476. Lawrence: University
Press of Kansas.
Velleman, J. David. 2006. Love as a Moral Emotion. In Self to Self: Selected Essays,
70–109. Cambridge (UK): Cambridge University Press.
Wallace, R. Jay. 2012. Duties of Love. Proceedings of the Aristotelian Society
Supplementary 86: 175–198. https://doi.org/10.1111/j.1467-­8349.
2012.00213.x.
214 N. L. ENGEL-HAWBECKER

Wedgwood, Ralph. 2006. The Normative Force of Reasoning. Noûs 40

(4): 660–686.
Whiting, Daniel. 2020. Aesthetic Reasons and the Demands They Do (Not)
Make. The Philosophical Quarterly pqaa045. https://doi.org/10.1093/
pq/pqaa045.
Wittgenstein, Ludwig. 1953. Philosophical Investigations. Translated by
G.E.M. Anscombe, P.M.S. Hacker, and Joachim Schulte, 4th ed. Wiley-­
Blackwell: West Sussex (UK).
Zangwill, Nick. 2013. Love: Gloriously Amoral and Arational. Philosophical
Explorations 16 (3): 298–314. https://doi.org/10.1080/1386979
5.2013.812739.
 CHAPTER 11

Love, Motivation, and Reasons: The Case

of the Drowning Wife

Monica Roland

11.1   Introduction
We commonly take ourselves to have good reasons for promoting the
interests and well-being of our loved ones. As a parent, for instance, you
do not question whether you have reasons to provide for your young chil-
dren on a daily basis, and as a friend, you do not check to see whether you
have reasons to support your friend in grief. Taking oneself to have rea-
sons for benevolent acts on behalf of loved ones thus seems to be an essen-
tial part of loving them. A more controversial issue in the philosophical
literature on love, however, is what the nature of those reasons is. For one
thing, there is disagreement about the source of those reasons. What
exactly provides us with reasons for benevolent acts on behalf of loved
ones? A different, but related, question is how these reasons relate to other
reasons for benevolent acts, such as moral reasons.
I address these questions here by considering Harry Frankfurt’s (2004)
and David Velleman’s (1999) respective accounts of love, with a particular
focus on their responses to one of the most well-known thought

M. Roland (*)
Oslo Metropolitan University, Oslo, Norway
e-mail: monicaro@oslomet.no

© The Author(s), under exclusive license to Springer Nature 215

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_11
216 M. ROLAND

experiments in moral philosophy, namely, Bernard Williams’s (1981)

hypothetical case of the drowning wife. This case is about a man who sees
two people on the verge of drowning, where one of them is his wife and
the other is a stranger, but where the man can save only one of them and
naturally wants to save his wife. Most responses to the case, including
Williams’s discussion of it, have primarily been concerned with the justifi-
catory basis for the husband’s preference and the particular difficulties
encountered by impartial moral theory in trying to legitimate his prefer-
ence. I will also be concerned with the husband’s reasons for saving his
wife over the stranger in this case. However, my aim here is not first and
foremost to contribute to the debate on impartial morality, but rather to
bring out what the case can tell us about love, in particular the relation
between loving someone and the reasons we have for acting on behalf of
those loved ones.
As a preliminary, I begin by outlining the main problems provided by
the case of the drowning wife. Next, I will present Frankfurt’s and
Velleman’s responses to the case and point out what I take to be problem-
atic about these responses. Both responses, I shall argue, neglect the inti-
mate relation between love and special relationships. I then proceed to
provide an argument for the claim that our reasons for benevolent acts on
behalf of loved ones are provided by the same things we essentially value
in love, namely, the inherent moral value of our beloveds and the special
relationships we have with them.1 I close with a discussion on special rela-
tionships and special reasons.

11.2   The Case of the Drowning Wife

Even though it was Williams’s discussion of the case that brought it to the
attention of a broader philosophical audience, the case of the drowning
wife first occurred in the work of Charles Fried (1970). Fried, who’s an

1
On my account, what we essentially value in love is not just the inherent moral value of
the beloved and the relationship one has with the beloved. When it comes to especially
romantic love and friendship, we also value the beloved’s laudable relational qualities (or
laudable moral character traits, broadly speaking). Mutual appreciation of such qualities is
constitutive of romantic relationships and friendships. However, the aim of this chapter is not
to account for the relation between what we value in love and the reasons for love, but rather
the relation between loving someone and the reasons lovers have for promoting the good of
the beloved. My focus here will thus not be on laudable relational qualities as such, but on
special relationships and the inherent value of the beloved. See otherwise my discussion on
reasons for love in my PhD thesis (2016), especially pp. 83–109.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 217

advocate of Kantian moral theory, points out that the requirements of

impartial moral principles seem to commit agents to always weigh the lives
of strangers and loved ones equally. This, he admits, provides the Kantian
with a problem in cases such as the one with the drowning wife, since it
would be absurd to insist that the husband should treat the wife and the
stranger equally, for instance, by flipping a coin in order to decide whom
to rescue.
As a solution to this problem, Fried suggests that as long as the hus-
band does not occupy any official position (i.e. as a lifeguard or the captain
of a ship), then “the occurrence of the accident may itself stand as a suffi-
cient randomizing event to meet the dictates of fairness, so he may prefer
his friend, or loved one” (227). He adds, furthermore, that if the husband
did occupy an official position, for instance, if he was the captain of a ship,
then “the argument that he must overlook personal ties is not unaccept-
able” (ibid.).
Williams was puzzled by Fried’s remarks on the husband’s reasons.
Here is the key passage in which he discusses the case:

[…] surely this is a justification on behalf of the rescuer, that the person he
chose to rescue was his wife? It depends on how much weight is carried by
‘justification’: the consideration that it was his wife is certainly, for instance,
an explanation which should silence comment. But something more ambi-
tious than this is usually intended, essentially involving the idea that moral
principle can legitimate his preference, yielding the conclusion that in situa-
tions of this kind it is at least all right (morally permissible) to save one’s wife
[…] But this construction provides the agent with one thought too many:
it might have been hoped by some (for instance, by his wife) that his moti-
vating thought, fully spelled out, would be the thought that it was his wife,
not that it was his wife, and that in situations of this kind it is permissible to
save one’s wife. (1981: 18)

Numerous philosophers have since commented on the case and dis-

cussed Williams’s remarks on the husband with “one thought too many.”2
Even though there are various ways of formulating the problems brought
out by the case, the debate seems in particular to have been centered on
three main questions. For one thing, the case raises the question of why
the husband is right to save his wife. That is, what justification does he

2
See, for example, Elinor Mason (1999), David Velleman (1999), Harry Frankfurt (2004),
Marcia Baron (2008), Troy Jollimore (2011, especially pp. 30–35), Susan Wolf (2012),
Simon Keller (2013), and Nicholas Smyth (2018).
218 M. ROLAND

have for his preference? This is what Nicholas Smyth (2018) has named the
justification problem. Smyth points out: “Ideally, [an answer to this ques-
tion] should not merely involve establishing that he is permitted to rescue
her, it also involves establishing that there are positive reasons in favor of
his action” (4). For another, one could ask why thoughts of justification
even should enter the husband’s mind in such a situation. Many consider
this to be the main question raised by Williams in the aforementioned
famous passage. Third and lastly, one could also ask how the reasons that
justify the husband’s preference are represented or integrated in the hus-
band’s actual motivations and dispositions. Smyth labels this the integra-
tion problem.
These questions will also serve as a backdrop for my discussion here, as
I proceed to explore the relation between our reasons for acting on behalf
of loved ones and the love we have for them.

11.3   The Husband with One Thought Too Many

So what exactly can the case of the drowning wife tell us about love and
our reasons for benevolent acts on behalf of loved ones? For one thing, it
tells us something about our expectations regarding the husband’s psycho-
logical makeup. Most responses to the case take for granted that the refer-
ence to the husband and the wife is meant to indicate that he loves her and
that he is compellingly motivated to save his wife over the stranger pre-
cisely because he loves her and not the stranger. Moreover, the case also
tells us something about our expectations regarding the explicit content of
the husband’s motivating thought. As we have seen, Williams argues that
any thoughts by the husband about justification for saving his wife over
the stranger surely would be “one thought too many.” On Williams’s
account, it would be somewhat inappropriate for the husband to look for
justification for his preference.
However, in an illuminating paper on the case, Susan Wolf points out
that “Williams does not himself call attention to the difference between
the case of the husband who thinks at the time of action about what is mor-
ally permissible for him to do and a different case in which one wonders,
retrospectively or counterfactually, what would be morally permissible and
why” (2012: 74). Nevertheless, she argues, most responses take for
granted that Williams talks about the former case, and the standard view
among those responses seems to be in agreement with Williams in that—
in this very case—reflection about permissibility at the time of the situation
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 219

would be one thought too many.3 It would be, Wolf argues, “something
unpalatable about a man who, faced with his drowning wife in one direc-
tion from his lifeboat and a drowning stranger in the other, checks to see
whether it is morally permissible before paddling (or diving) toward his
wife” (ibid.). Just as a lack of motivation to save the wife seems incompat-
ible with love, so it seems, is reflection in the heat of the situation about
permissibility to save the wife over the stranger. We expect some sort of
automaticity in this particular case and automaticity leaves no room for
that sort of reflection.
Let me note, however, that this is not unique for love. If there were just
one person out in the water—a stranger—most of us, I believe, would
think that it would be one thought too many also for the moral agent if he
had to reflect about whether he has a reason to rescue the drowning
stranger. It would be something unpalatable also about someone who had
to check whether he should save a stranger from drowning, or whether it
is permissible not to save the stranger and just walk away. We expect some
sort of automaticity also in cases that do not involve love as such, that is,
cases in which we do not love the other person. Automaticity might be
expected in many, perhaps even most, cases involving loving agents, but it
does so, I believe, also for many cases involving merely moral agents.
Nevertheless, even if we expect of the husband that he does not think
about justifying reasons at the time of action, that, in fact, thinking about
reasons in the heat of the situation would be incompatible with him loving
his wife, many philosophers still want to provide justification for his par-
tiality. That is, it is one thing to oppose the idea that the husband should
think of reasons at the time of the situation, another to oppose the idea
that there are reasons that legitimate the husband’s preference. Thus, if
the husband should save his wife, what is the justification for him doing

3
In response to Wolf’s claim here, Nicholas Smyth argues that Williams clearly talks about
permissibility at the time of the situation. He says, “Though some of his defenders have
sought to smooth over this fact, his references to the husband’s ‘motivating thought’ make
it very clear that Williams was specifically worried about how Kantian theory will require us
to think and feel when we are acting” (2018: 4). Even though I agree with Smyth’s observa-
tion here that Williams was talking about the husband’s thoughts at the moment of the situ-
ation, I still think Wolf has a valid point. When we talk about the husband’s reasons in this
case, there is an important difference between (1) the husband’s motivation and thoughts at
the time of action and (2) thinking about justification for his preference retrospectively (or
counterfactually).
220 M. ROLAND

so? What provides him with overriding reasons for rescuing the wife rather
than the stranger?
I will now turn to the ways in which Frankfurt and Velleman have
responded to the case. Frankfurt and Velleman offer two different solu-
tions to the aforementioned questions: While Frankfurt argues that the
husband’s reason for saving his wife over the stranger is given by his love
for her, Velleman insists that the husband’s reason for saving his wife in
this case has nothing essentially to do with love, but is rather provided by
their relationship.

11.4   Frankfurt’s Response

Consider these passages in which Frankfurt reflects on the famous thought
experiment:

[…] the example as [Williams] presents it is significantly out of focus. It can-

not work the way he intends, if what is stipulated concerning one of the
drowning people is merely that she is the man’s wife. After all, suppose that
for quite good reasons the man detests and fears his wife. Suppose that she
detests him too, and that she has recently engaged in several viciously deter-
mined attempts to murder him. Or suppose that it was nothing but a cold-­
bloodedly arranged marriage of convenience anyhow, and that they have
never ever been in the same room together except during a perfunctory
two-minute wedding ceremony thirty years ago. Surely, to specify nothing
more than a bare legal relationship between the man and the drowning
woman misses the point. (2004: 36–37)

If he does truly love her, then he necessarily already has that reason. It is
simply that she is in trouble and needs his help. Just in itself, the fact that he
loves her entails that he takes her distress as a more powerful reason for
going to her aid than for going to the aid of someone about whom he
knows nothing. The need of his beloved for help provides him with this
reason, without requiring that he think of any additional considerations and
without the interposition of any general rules […] If the man does not rec-
ognize the distress of the woman he loves as a reason for saving her rather
than the stranger, then he does not genuinely love her at all. Loving some-
one or something essentially means or consists in, among other things, tak-
ing its interests as reasons for acting to serve those interests. Love is itself,
for the lover, a source of reasons. It creates the reasons by which his acts of
loving concern and devotion are inspired. (37)
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 221

As Frankfurt rightfully points out here, neither bare legal relation-

ships—for instance, the mere fact that two people are married—nor unlov-
ing or hateful relationships do generally provide reasons for partisan
actions on behalf of the other. However, Frankfurt does not consider the
possibility that loving relationships can provide reasons for partiality in
action, and I find it a bit surprising that he does not address that issue
specifically or in more detail, since that should be our starting point.4
As already argued, we should take for granted that the husband and the
wife in this very case do share a history of mutual concern and love for one
another. The reference to their relationship is thus not just a reference to
their legal relationship, but also to their historical relationship and the lov-
ing character of it, and thus not irrelevant for the case. A loving relation-
ship is a very real thing in the world, and it seems very plausible that loving
relationships are normatively significant. Why shouldn’t they be? As, for
instance, Niko Kolodny (2003) points out: If you are waiting for the out-
come of your mother’s surgery, and a hospital volunteer asks you why you
are so worried about this patient in particular, then the natural reply would
be: “Because she’s my mother.” Reference to the relationship provides a
reason that makes your behavior at least as intelligible, permissible, and
appropriate as if you said, “Because I love her.” On Frankfurt’s account,
however, it seems that only the latter reply refers to a reason for your
behavior. I think that is a problematic view.
That being said, I agree with Frankfurt that the husband qua lover does
not need to reflect in the heat of the moment about what reasons there are
for saving his wife in order to be compellingly motivated to do so or for
generally being aware of reasons for such an act. As argued earlier, his love
for his wife explains his motivation for saving her in particular, and if asked
later on (after the incident) he will surely have access to reasons for rescu-
ing her. But I see no argument on Frankfurt’s behalf for why that neces-
sarily excludes the possibility that loving relationships can be normatively

4
One could argue, in Frankfurt’s defense, that loving relationships might be what he has
in mind when objecting to the idea that reference merely to the fact that she is the man’s wife
misses the point, since that fact does not reveal anything about the quality of the relationship
or whether they love each other. Thus, he could be understood as saying that not any type of
relationship provides reasons, merely loving or benevolent relationships. However, in the
remainder of his discussion on Williams’s thought experiment, he does not refer to loving
relationships as reasons, nor does he say anything about the relation between relationships
and love. Rather, Frankfurt refuses to give relationships a prominent role in his account
of love.
222 M. ROLAND

significant. Rather, as already suggested, the absence of explicit reflection

on reasons at the time of the situation does not need to have any bearing
on what reasons there are for preference in this case. Furthermore, auto-
maticity in such cases might in itself be a manifestation of internalized
reasons that govern an agent’s actions without it being the case that he
consciously reflects upon them when he acts.
The problem, as I see it, with Frankfurt’s response is thus twofold. For
one thing, he rejects the possibility that loving relationships are norma-
tively significant, without providing a good argument for such a rejection.
This is, of course, not in itself an argument against Frankfurt’s positive
claim that love provides reasons for partiality, but it is an argument against
accepting his negative claim—his rejection of special relationships, such as
spousal relationships, as legitimate sources of partiality—without further
inquiry. For another, Frankfurt ignores the constitutive role lovers’ valua-
tion of their relationships with their loved ones have for love. Loving
someone, I argue, is, in part, to value the relationship you have with that
person. This, again, might have implications for how we should under-
stand the claim that love creates reasons.

11.5   Velleman’s Response

Velleman has a different response to Williams’s hypothetical case from
Frankfurt’s. In his account, the husband’s reasons for saving his wife over
the stranger have nothing essentially to do with his love for her. He says:

Of course the man in Williams’s story should save his wife in preference to
strangers. But the reasons why he should save her have nothing essentially
to do with love […] The grounds for preference in this case include, to
begin with, the mutual commitments and dependencies of a loving relation-
ship. What the wife should say to her husband if he hesitates about saving
her is not “What about me?” but “What about us?” That is, she should
invoke their partnership or shared history rather than the value placed on
her by his love. Invoking her individual value in the eyes of love would
merely remind him that she was no more worthy of survival than the other
potential victims, each of whom can ask “What about me?” (1999: 373)

I have several comments on Velleman’s response here. The first com-

ment concerns the source of reasons in normal cases of benevolent acts on
behalf of loved ones, the second comment concerns the source of reasons
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 223

in this particular case, and the third concerns the relation between love
and loving relationships. However, in order to bring some context to the
aforementioned quote and the following discussion, I will first say a few
words about Velleman’s account of love.
According to Velleman, love is a moral emotion, and it is moral in the
sense that it necessarily involves the moral attitude of respect. Now, in ordi-
nary language the word “respect” has a number of different, though
related, meanings. For instance, it might express a sort of appraisal or
admiration for someone in virtue of their character, skills, or achieve-
ments.5 However, the word can also refer to a type of attitude we should
have toward all persons regardless of their merits, social status, and char-
acter. All persons, it seems, are entitled to a certain type of respect simply
in virtue of being persons. This idea grounds most contemporary thought
in moral philosophy, and it is this meaning of the word that lies at the
heart of Velleman’s account as well.6 For Velleman, love and respect alike
are responses to a value we all share, namely our inherent moral value as
ends in ourselves. However, whereas respect is an attitude we should have
toward everyone in virtue of this value, there are no such moral require-
ments on love. He says, “I regard respect and love as the required mini-
mum and optional maximum responses to one and the same value” (366).
One implication of the view that love and respect alike are responses to
the inherent value of persons is that both have significant impact on our
wills (i.e., on our motivations and dispositions). In the Kantian framework
Velleman assumes, ends in themselves have value for their own sake, and
the recognition of this inherent value of others forces us to treat them as
sources of valid claims. For respect, the impact of such recognition on our
wills is predominantly negative; it constrains what we can do to other
people; it arrests our egoistic inclinations to use others as mere means for
our own ends. But respect can also be a source of positive motivation and
duties, such as in situations where one is in a position to save strangers
from drowning. For the case of love, however, the impact on our wills is
not primarily negative, though it certainly is an essential part of love that

5
The example is in no way meant to be an exhaustive analysis of different types of respect
or of the various uses of the word in natural language. For a more detailed discussion on the
matter, see, for instance, Stephen L. Darwall (1977), Carla Bagnoli (2006), and the intro-
ductory section of John J. Drummond (2006).
6
Thus, it is possible to lose respect for someone, in the sense that one no longer admires
their moral character or respects their political views, but still has respect for them in the
sense that one acknowledges their inherent moral value and thus their humanity.
224 M. ROLAND

one does not use the other as mere means to pursue one’s own interests.
Rather, according to Velleman, love is to a greater extent than mere respect
a source of positive motivation in that it results in a “heightened sensitivity
to the other’s interests” (361). Loving someone involves a motivational
disposition to promote their interests and well-being—in some cases on a
daily basis. With this in mind, we can now look at some of the implications
of Velleman’s view.
The source of reasons in normal cases of benevolent acts on behalf of loved
ones: For one thing, it seems that Velleman is committed to the view that
the basic reason for benevolent acts on behalf of any person—loved ones
or strangers—derives from their inherent moral value. The inherent value
of others provides us with basic reasons for acting well toward them, in
particular when such acts are needed. Thus, for the husband in Williams’s
hypothetical case, the basic reason for saving his wife is the same as the
basic reason for saving the stranger; it is grounded in their inherent value
as ends in themselves. This, of course, lies at the heart of the very problem;
their inherent value cannot provide justification for preference. Velleman’s
caution against invoking the wife’s individual value reflects this very point.7
Second and following from this, Velleman thus seems committed to the
view that one’s basic reason in general for acting on behalf of loved ones
essentially has something to do with love. In normal cases, what we essentially
value in love, according to Velleman, namely the inherent value of the
beloved, is also the source of our basic reason for acting well toward them.
Lovers’ basic reasons for promoting the good of their beloveds are
grounded in their beloveds’ inherent moral value.
Third, it seems, however, that the inherent moral value of our beloveds
is not the only significant normative factor in cases concerning benevolent
acts on their behalf. If special relationships can provide reasons for partial-
ity in cases such as the one with the drowning wife, I see no reason why
they cannot do so in normal cases as well. For one thing, being in a loving
relationship with someone will give you unique insight into the needs and
interests of that particular person, as well as the opportunity and perhaps
even skill to promote their interests specifically. This seems to provide you
7
Following Kant, Velleman holds the view that the inherent value of persons is different
from the value of other things in that it does not allow for comparisons among alternatives.
Persons have the value of dignity, while other things have the value of price, which do allow
for comparisons. Thus, in cases where one has to choose between saving one person over the
other, the decision of whom to choose must be made on the basis of other factors than the
incomparable value of dignity that they both share.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 225

in particular with reasons to promote such goods. Velleman would pre-

sumably not object to the claim that special relationships provide lovers
with reasons for benevolent acts on behalf of loved ones also in normal
cases. However, based on his comment in the aforementioned quote, he
still seems to reject the claim that lovers’ appreciation of such relationships
is part of the very attitudes that constitute their love. I think that is an odd
position and I will explore this point in more detail.
The source of reasons in Williams’s thought experiment: Velleman seems
to think that this case differs from normal cases of benevolent acts on
behalf of loved ones, on the grounds that in this case what is valued in love
and the reasons for acting on the motives of love come apart. It is not the
wife’s inherent value that provides the husband with reasons for saving
her, Velleman stresses, but their relationship. I think this claim is in need
of clarification.
For one thing, Velleman’s wording in the aforementioned quote—“the
reasons why he should save her have nothing essentially to with love”—is
inaccurate, or at least ambiguous. If what I have argued earlier is correct,
then it follows that Velleman is committed to the view that the basic rea-
son for saving the wife has something to do with love. Both love and the
basic reason for saving her are grounded in her inherent value. It is the
reason for preference that according to Velleman has nothing to do with
love. That is, if love is merely a response to the beloved’s inherent moral
value, then referring to their spousal relationship as a source of reasons for
preference is a reference to something external to love.8
However, as I have already argued, there are reasons to think that love
is not merely a response to the beloved’s moral value. If the husband in
this case considers their spousal relationship a reason for saving her over a
stranger with whom he does not share such a relationship, it follows that
he values that very relationship. Taking the relationship to provide such a
reason is to value it. Furthermore, as I will argue for in more detail later,
his valuation of the relationship seems to be an essential part of his love for
his wife. It is his relationship with his wife—understood as their shared

8
This is precisely the problem for impartial morality, as well. If the reasons for moral acts
on behalf of others are grounded in their inherent value and nothing else, then it seems that
morality cannot help us decide whom to choose in this case, just that we are morally required
to rescue one of them or as many as we can. If Velleman holds such a view, then it seems that
on his account both love and morality fall short in providing justification for saving the wife
over the stranger in this case. This might have been what Williams had in mind when he said,
“some situations lie beyond justifications” (1981: 17).
226 M. ROLAND

history together—that makes her irreplaceable to him. On these grounds,

I am thus skeptical to the claim that lovers’ appreciation of their relation-
ships with their loved ones is not part of the very attitudes that constitute
their love and thus, by extension, that the source of justification for prefer-
ence in this case comes apart from what is valued in love. This brings me
to my last comment in this section regarding Velleman’s response to
Williams’s thought experiment.
The relation between love and loving relationships: The question of what
we essentially value in love—that is, what love essentially is a response
to—is hotly debated in the philosophical literature. Just like Velleman, I
hold the view that love is a response to the inherent moral value of the
beloved. Note that this is not a claim about causality, and thus what elicits
love. Rather, it is a claim about one of the essential attitudes of love. Love,
I argue, necessarily involves the moral attitude of respect. You cannot love
someone without acknowledging their value as ends in themselves. This
implies for one thing that when we love someone our interest in the other
is not primarily instrumental. Rather, we care about the person as some-
one that is important for their own sake. Furthermore, I also believe that
recognition of the beloved’s moral value involves taking that value to be a
source of basic reasons for benevolent acts on their behalf.
However, unlike Velleman, I argue for what we could label a “dual
account of love.” Loving someone is not merely a valuation of a generic
value we all share, but also a valuation of particulars, such as the particular
relationship one has with the beloved. After all, it is our relationships with
our loved ones that make them special to us. Thus, if love is essentially
selective, it must be reflected in its constitutive attitudes. It cannot just be
a valuation of a generic feature, but also of features that distinguish the
beloved from others, such as the special relationship one has with the
beloved. Moreover, valuing the relationship one has with a beloved
involves seeing that relationship as a source of special reasons for acting for
the good of that person—reasons that typically do not apply to others who
are not in such a relationship with one’s beloved.
On these grounds, I thus take issue with what seems to me to be an
artificial distinction in Velleman’s account between love and the lover’s
valuation of the relationship. Velleman lacks an argument for why the hus-
band’s valuation of the relationship he has with his wife is a different psy-
chological phenomenon than the love he has for her.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 227

11.6   The Possibility of a Moral Account of Love

In the earlier discussion, I have suggested that what we essentially value in
love is both the inherent moral value of the beloved and the special rela-
tionship one has with the beloved. Furthermore, I have also suggested
that our reasons for beneficial acts on behalf of loved ones are provided by
the same things we value in love. The inherent moral value of the beloved
provides the lover with basic reasons for beneficial acts on his or her behalf
and the special relationship one has with the beloved provides the lover
with special reasons for such acts. This duality of love suggests that love, at
least in part, is a moral emotion, and furthermore, that our basic reasons
for good deeds on behalf of loved ones have a moral basis.
While Velleman argues for a moral account of love, Frankfurt is more
resistant to the idea that love is inherently moral. This skepticism rests on
his view that love is not a result of moral reasoning or an awareness of a
moral value in the beloved, and, furthermore, his view that love and
morality are two different sources of normativity. For one thing, Frankfurt
points out: “love does not require a response by the lover to any real or
imagined value in what he loves […] Love is not a conclusion. It is not an
outcome of reasoning, or a consequence of reasons. It creates reasons”
(2006: 25). For another, he argues, even though “the grip and the force-
fulness of the requirements that love imposes upon us resemble the force-
fulness and the grip of the demands that are made upon us by moral
obligation,” it is still the case that “the necessities that characteristically
grip us in the one sort of case have different grounds than the necessities
that characteristically grip us in the other” (1998: 5).
Frankfurt is arguably right when he contends that love does not typi-
cally arise as a result of reasoning. Most instances of love are not direct
results of a decision to love, though there might be cases where someone
has come to love another person after realizing they have good reasons for
doing so. Both Berit Brogaard (2015) and S. Matthew Liao (2007), for
instance, argue that over time and with effort, one can elicit emotions one
judges as appropriate and desirable.9 However, I am more skeptical of

9
Consider, for instance, the case of a parent who has trouble connecting emotionally to his
or her newborn child, but who acknowledges that s/he has strong reasons to care for and
love the child and then does everything in their power to develop proper love for the child.
This seems to me to be a case in which the parent’s love for their child is a consequence of
reasons in the sense that awareness of those reasons is what drives the parent in their effort
to achieve the desired emotional response.
228 M. ROLAND

Frankfurt’s rejection of love as a response to the value of the beloved, as

well as his claim that the reasons of love and morality necessarily have dif-
ferent grounds. More precisely, I argue that his account of love as a disin-
terested concern for the beloved is in tension with his non-moral account
of love. Let me explain.
On Frankfurt’s account, love is a disinterested concern for the beloved
and his or her well-being.10 When we love someone, he argues, our inter-
est in the other is not essentially instrumental; it is not grounded in a
desire for some other goal, nor is it grounded in mere self-interest.11
Rather, we care about the beloved essentially as an end. He says, “It is in
the nature of loving that we consider its objects to be valuable in them-
selves and to be important for their own sakes” (2006: 42). The upshot of
this disinterested concern is twofold. For one thing, Frankfurt argues, love
is “a configuration of the will that consists in a practical concern for what
is good for the beloved” (43). Love turns the lover’s attention to the
beloved’s well-being and interests, and promoting those interests becomes
part of the lover’s own interests. For another, love imposes certain con-
straints upon our wills. On Frankfurt’s account, love is a kind of volitional
necessity. Loving someone entails that there are certain things we feel we
must do for the beloved and certain things we cannot bring ourselves to
do. As lovers, Frankfurt argues, we cannot help being guided by these
constraints and motives.
However, to consider someone to be important for his or her own sake
is not just essential to love, but also to respect. That is, if love essentially is
a disinterested concern for the beloved, and if this disinterested concern
consists in valuing that person as an end in him- or herself, then love seems
to share much of its features with the moral attitude of respect. By exten-
sion, if one of the upshots of love being such a disinterested concern is that
it constrains our wills in that it prevents us from treating the beloved
essentially as means to some further end, then the volitional constraints
essential to love also seem to have something important in common with
the moral constraints that respect imposes upon our wills. As we have

10
For similar claims on the disinterested character of love, see, for instance, Niko Kolodny
(2003), who suggests that love is a non-instrumental valuing of the beloved, and Kate
Abramson and Adam Leite (2011), who argue that love is a non-self-interested response.
11
Even if love essentially is a disinterested concern for the other, it does not follow that we
do not also value our loved ones interestedly. Insofar as it is important for the lover to have
her beloved in her life, she is also valuing her interestedly and thus as instrumental for her
well-being. See, for instance, Susan Wolf (2012: 85) for a similar claim.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 229

seen, this is precisely Velleman’s view. It is also my view. We both argue

that love involves the moral attitude of respect.
Frankfurt does not say much about the relation between love as a dis-
interested concern and respect. However, it is hard to see how caring
about someone as an end would not involve a moral attitude toward that
person. In fact, if a disinterested concern for a particular person does not
involve the moral attitude of respect, it is hard to understand what caring
about someone as an end even means. In Frankfurt’s defense, it could be
that he uses the notion as an attempt to contrast love from egoistic drives,
which per definition are interested. Still, as far as I can tell, he does not
offer an argument for why the disinterested concern essential to love is not
a moral attitude. It is unclear therefore whether, on Frankfurt’s account,
there can be (disinterested) love without respect. If yes, then what distin-
guishes this type of disinterested concern from the disinterested concern
essential to respect? Frankfurt does not say. If no, then it seems Frankfurt’s
account of love as essentially disinterested conflicts with his non-moral
account of love. Caring about someone as ends in themselves is to attri-
bute them moral value.
One reading of Frankfurt’s account of disinterested love thus opens up
the possibility that love involves the moral attitude of respect. This is in
any case the view I defend. What love and morality have in common is, for
one thing, that the volitional constraints both impose upon us have the
same ground; they are grounded in the value of the other as an end in
him- or herself. The volitional constraints of love that prevent one from
treating the beloved merely as means to some other end are also moral
constraints. Even if these volitional constraints might have different moti-
vational sources in the sense that some of them play out in the context of
loving relationships and some of them do not, they still share a common
ground: the other’s inherent value. For another, even though love to a
greater extent than mere respect is a source of positive motivation, the
positive acts that both promote are also grounded in the inherent value of
the other and the recognition that the other is important for his or her
own sake, and thus a source of valid claims.

11.7   A Too Moralistic Account of Love?

The skeptic might argue that an account of love as necessarily involving
the moral attitude of respect is both implausible and too moralistic. Real
life, real people, and everyday relationships do not work like that. We do
230 M. ROLAND

not always treat each other with respect—we sometimes say and do things
to harm those closest to us; we sometimes act selfishly or even childishly.
However, this occasional failure to act respectfully toward our loved ones
does not mean, as my account seems to imply, that we do not love them.
It just means that intimate relationships are more messy and complicated
than idealized philosophical theories account for.
To be clear, my claim is not that loving someone is a warrant for not
making mistakes, that if we love someone we will never act in ways that
will hurt them. We are fallible creatures. However, as Samuel Scheffler
points out, “to say that we are fallible is not to say that we are systemati-
cally misguided” (2010: 106). We are creatures with the capacity to per-
ceive others as ends in themselves, and we typically relate to others as such
ends. One of the ways in which we relate to others as ends in themselves is
by loving them—and loving someone essentially involves a general dispo-
sition to promote their well-being and flourishing. As lovers we cannot
help being such motivated. Even if there may be occasions where we fail
to provide proper care and attention, too much failure of the sort will be
a failure to love. It will simply not be intelligible as a case of love.
This, I believe, harmonizes well with our common intuitions about
love. We commonly make normative judgments about love and intimate
relationships. Consider, for instance, our responses to abusive and thor-
oughly destructive relationships; we deem such relationships unloving and
describe them as harmful and not the way love should be. Love is just not
compatible with that sort of moral violation.
On that note, I will now turn to special relationships. What are they and
how do they provide lovers with special reasons for promoting the good
of their beloveds?

11.8   Special Relationships

It is undeniably true that our loved ones have a special value to us, a value
we do not attribute to others whom we do not love. Frankfurt describes
the relation between loving someone and the value we place on them by
referring to the love he has for his children. He says, “The reason they are
so precious to me is simply that I love them so much […] it is plainly on
account of my love for them that they have acquired in my eyes a value that
otherwise they would certainly not possess” (2004: 40).
I obviously agree with Frankfurt that love necessarily makes the beloved
valuable to the lover, but his use of the notions “simply” and “plainly”
seems to imply the even further claim that our loving someone is the only
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 231

(relevant) thing that accounts for the special value of our beloveds to us.
But such a view merely tells half the story. Even if it is trivially true that we
would not value our beloveds in the same way if we did not love them
(love, after all, is a type of valuing), it does not follow that our loving them
is the only thing that accounts for the special value they have for us. That
is, it is just as true to say that what makes our loved ones so special to us
are the special relationships we have with them. For instance, if the per-
sons in question in Frankfurt’s case were not his children, then they would
not have had that particular value to him. Just the same, if it were not for
the existence of the special relationship between two of the people in
Williams’s hypothetical case, it is unlikely that the rescuer would have felt
such a strong preference for rescuing one of the drowning persons rather
than the other.12 Relationships matter; they are vital for the development
of love and thus for the particular value we place on our beloveds.
This perspective, I have argued, is lacking or underdeveloped in both
Frankfurt’s and Velleman’s accounts of love. None of them offers a satis-
factory account of the selectivity of love: of why we love the particular
persons that we do, why we have reasons for such selectivity, and how
appreciation of these reasons for selectivity is an essential part of love.13

12
Granted, we could think of a hypothetical case where the rescuer would feel a strong
preference for rescuing one over the other without it being the case that he loved the person
in question. For instance, the rescuer might have no personal ties to any of the drowning
persons, but still know that one of them is a mass murderer and the other the inventor of the
Covid-19 vaccine. In such an unlikely situation, the rescuer would probably have a strong
preference for saving the inventor of the vaccine.
13
Admittedly, Frankfurt addresses the issue of selectivity, but he does not relate the selec-
tivity of love to reasons for love or the impact those reasons can have on the lover’s motiva-
tions and dispositions. He says: “The reason it makes no sense for a person to consider
accepting a substitute for his beloved is not that what he loves happens to be qualitatively
distinct. The reason is that he loves it in its essentially irreproducible concreteness. The focus
of a person’s love is not those general and hence repeatable characteristics that make his
beloved describable. Rather, it is the specific particularity that makes his beloved nameable—
something that is more mysterious than describability and that is in any case manifestly
impossible to define” (1999: 170). I am uncertain of what this means. As a comment to this
passage, Kolodny points out that “the beloved’s bare identity […] cannot serve as a reason
for loving her. To say ‘She is Jane’ is simply to identify a particular with itself. It is to say
nothing about that particular that might explain why a specific response is called for” (2003:
142). I partly agree with Kolodny. To say, “She is Jane” does not in itself explain why love
for Jane is an appropriate response. However, if Jane were one of Frankfurt’s children, then
the proposition “She is Jane” would, for Frankfurt, entail reference to the relationship he has
with her and thus explain why she has the special value she has for him. However, I do not
think that this is what Frankfurt has in mind when referring to that which makes the beloved
nameable.
232 M. ROLAND

None of them offer an account of special (or loving) relationships, though

admittedly, Velleman briefly addresses the importance of the relational
context for love in the introduction to his anthology Self to Self: Selected
Essays (2006). Here he says: “Personal love is an essentially experiential
emotion: a response to someone with whom we are acquainted. We may
admire or envy people of whom we have only heard or read, but we can
only love the people we know” (10–11).
I believe Velleman is right here. Interacting with and getting to know
someone is necessary for developing love for them, and relationships pro-
vide contexts in which such interaction can take place. Moreover, the exis-
tence of special relationships, such as romantic relationships, close
friendships, and familial relationships, makes participants’ love for each
other appropriate responses, whereas the absence of a special relationship
would make love inappropriate. For instance, it would make no sense if a
complete stranger claimed to love Ben; in fact, it would seem out of place
and deluded, whereas the fact that Tom is Ben’s father makes Tom’s love
for Ben both intelligible and appropriate.14 Tom’s love for Ben makes
sense both for Tom and Ben, and it also makes sense for us as bystanders.
Special relationships account for the special value our beloveds have for us.
It is not a mystery why they have become so precious to us.
Niko Kolodny is perhaps the most prominent proponent of the “rela-
tionship view,” the view that special relationships provide normative rea-
sons for love and its constitutive motivations, and, furthermore, that
loving someone partly consists in valuing the relationship one has with the
beloved. According to Kolodny (2003), special relationships can both
account for the selectivity of love and the appropriateness of such a selective
love. That is, special relationships provide us with reasons to love some

14
The term “complete stranger” is meant to imply that neither Ben nor the person claim-
ing to love Ben has ever met or interacted with each other before or even heard of each other.
However, there are real-life cases, where persons claim to love someone with whom they do
not share a personal relationship and where the beloved does not even know that the other
exists, but where their love seems to be genuine and not an inappropriate response—though
perhaps of a different kind than the personal love I address in this chapter. Consider, for
instance, the love that many Brits seemed to have for Princess Diana. Many felt they knew
her, and one could say that there was a loving relationship between the people and the prin-
cess. There was a history of interaction and mutual concern between them, even though the
relationship was more asymmetric than in cases of personal love. While individual persons
loved the princess, her love was not directed at them as individuals, but as a group. I will not
pursue this topic any further here.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 233

people over others and they also provide us with reasons to love them in
particular ways. How so?
On Kolodny’s account, special relationships are constituted by the
ongoing history of interaction and mutual concern participants have for
each other.15 It is these shared histories that explain why we end up loving
the particular persons we do, and not others with similar personal qualities
but with whom we do not share such histories.16 Furthermore, these his-
tories of mutual concern and interaction also make love an appropriate or
fitting response. Just as, for instance, fear is a fitting response to dangerous
situations, so is love for one’s beloved a fitting response to the reality of
the relationship. Special relationships justify that one loves, say, one’s
friend and not a random stranger.
Still, Kolodny points out, appropriateness is not just about whom we
should love, but also about how we should love. We expect friends, for
instance, to love us differently from our parents. Thus, the fact that Jane
is my friend (that I have that type of relationship with her) is not just a
reason for me to love her in particular, but also a reason for me to love her
in the way fitting for friendship, and not, say, maternal love. I think this is
an important point and a perspective that is missing from both Frankfurt’s
and Velleman’s accounts. When we love someone, we do not just respond
to their inherent value as persons, but also to certain relational features.

15
This is a simplification of his view. According to Kolodny, there are two categories of
special relationships: (i) romantic relationships and friendships are necessarily historical; they
cannot exist without a history of interaction and mutual concern between its participants,
while (ii) there is a sense in which familial relationships can exist without such a shared his-
tory. Even if a history of interaction and mutual concern ideally and typically characterizes
close family relationships, one can be in a family relation with someone even if those condi-
tions are not met. Think about the father, for instance, who finds out that he has a child he
did not know about. However, this difference in necessary conditions between familial rela-
tionships on the one hand, and romantic relationships and friendships on the other, is not
important for our discussion here. I will merely take as my starting point that, on Kolodny’s
account, a shared history of mutual concern and interaction is essential for special relation-
ships and the development of love.
16
This is also known as the problem of substitution: If love is a response to someone’s
favorable qualities, such as their blue eyes and wit, it seems to imply that anyone with the
relevantly similar features could or should be a substitute. In the same manner, if we love our
spouses, parents, siblings, children, and friends in response to their inherent moral value—a
value we all share—then it seems just as appropriate to love a stranger in response to his or
her inherent moral value. The “relationship view” avoids the problem of substitution. For
discussions on the problem of substitution, see, for example, Niko Kolodny (2003), espe-
cially p. 141, and Bennett Helm (2010), especially pp. 24–25.
234 M. ROLAND

For instance, it seems essential to my love for my brother that I do not just
value him as a person, but that I also value him as my brother. The kind of
love I have for him—sibling love—and thus the special value he has for
me, reflects the fact that he is my brother.
What is further appealing about Kolodny’s account is his emphasis on
how awareness of the fact that one is in a special relationship with the
beloved figures in the constitutive attitudes of love. Love, he argues, “is
not only rendered normatively appropriate by the presence of a relation-
ship. Love, moreover, partly consists in the belief that some relationship
renders it appropriate, and the emotions and motivations of love are caus-
ally sustained by this belief” (2003: 146). According to Kolodny, the lover
sees the relationship as a reason to value both the beloved and the rela-
tionship: “love consists (a) in seeing a relationship in which one is involved
as a reason for valuing both one’s relationship and the person with whom
one has that relationship, and (b) in valuing that relationship and person
accordingly” (150).
Now, I do not agree with Kolodny on every aspect of his account. For
instance, we disagree about the reasons for romantic love and friendship
and thus what constitutes the very relationships that ground these types of
love. That is, on my account, romantic relationships and friendships are
not just constituted by histories of interaction and mutual concern between
its participants, but also by participants’ mutual appreciation of each oth-
er’s laudable relational qualities (or laudable moral character traits), for
instance, kindness. When it comes to friendship and romantic love, I do
not see how a relationship theory could even make sense if it did not view
lovers’ mutual appreciation of each other’s relational qualities as partly
constitutive of the relationship. However, this disagreement does not have
implications for our discussion here. Even if appreciation of laudable rela-
tional qualities is important for the development and appropriateness of
romantic love and friendship, once the relationship is established, it pro-
vides lovers with reasons for benevolent acts on behalf of their loved ones.
The relationship view, then, has the advantage that it makes it intelli-
gible why we love the ones we do, and it also provides a plausible account
of the lover’s psychology. We all know in a general way why our love and
acts of love are appropriate. As a lover, one is aware that one is in a special
relationship with the beloved and that this special relationship warrants
the response of love.
Having explored the role of special relationships, I will end this chapter
with a few comments on special reasons.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 235

11.9   Special Reasons

The inherent moral value of persons makes claims on all of us (or on all
moral agents). We all have reasons to act well toward others in virtue of
the inherent value persons possess as ends in themselves. I have called
these reasons basic reasons.17 However, as lovers, we have even further
reasons to act well toward our loved ones. As lovers, it is not just the
inherent moral value of our beloveds that makes claims on us, but also the
special relationships we have with them. Special relationships and the
mutual expectations and commitments that follow from them, commit
the lover specifically to promote the good of the beloved. So how should
we think about these special reasons and their relation to special
relationships?
When we act well toward others we provide goods to the persons we
act on behalf of. Simon Keller (2013) points out that there seem to be two
types of goods: Generic goods “are goods we could in principle provide to
anyone,” while special goods “can be provided only within the context of a
particular relationship” (111). There is “only one, or only a few people by
whom it can be provided” (106). Imagine, for instance, that your close
friend just lost her partner to cancer. Her need for emotional support and
comfort and the fact that you are her closest friend, provide you with a
strong reason for providing her with that support and comfort. The emo-
tional support you can provide in virtue of being her long-time friend is a
special good that cannot be provided by just anyone. For one thing, very
few know her as well as you do and are as familiar with her needs and
vulnerabilities.
Imagine, next, that a stranger hears about the death of your friend’s
spouse and shows up at her house offering her a shoulder to cry on. Such
an act would probably come across as invasive and inappropriate. Special
relationships sometimes make a difference to what one is allowed to do.
Certain acts that would provide goods if they were performed by loved
ones may not provide goods, but rather the opposite, if performed by
strangers. Special relationships, then, not only put us in a unique position,
both causally and epistemically, to promote the goods of our loved ones,
but they also make our acts of love appropriate. Both you and your
17
In the philosophical literature, these reasons are often called agent-neutral reasons, a
notion that is focusing on (the duties of) the acting agent and not as much on (the rights of)
the person acted on. I use the term basic reasons here to emphasize that these reasons are
provided by the inherent value of the person for which these good deeds are done.
236 M. ROLAND

mourning friend are aware of the special reasons (and expectations) pro-
vided by your loving relationship. In the absence of the right sort of expla-
nation, a lack of loving support in this situation would not only constitute
grounds for criticism, but also seem incompatible with love.
Generic goods, however, are not in the same way conditional upon the
relationship. Imagine, for instance, a version of Williams’s thought experi-
ment in which there is just one person on the verge of drowning, namely
your wife, and that there are two people in a position to save her, you and
a stranger. Needless to say, you both have very strong reasons for saving
her. You are both morally required to do so. Furthermore, the generic
good achieved by her being saved does not depend on whether or not she
has a special relationship with her rescuer. This is a good that in principle
could be provided by anyone. Still, the fact that it is your wife out in the
water, and not a stranger, seems to provide you with an additional reason
the other rescuer does not have. How should we think about such a reason?
On the one hand, both you and the stranger have a basic reason for
rescuing your wife, and this reason is grounded in her inherent moral
value (and, of course, the fact that her life is at stake). For another, this
reason can also be described as an agent-neutral reason for rescuing her,
given that your reason for rescuing her is not dependent on having a per-
sonal relationship with her. It is safe to say that you both have sufficient
reason for saving her. You do not need an additional reason. However, it
seems that you do have an additional reason in virtue of being her spouse.
Your relationship with your wife provides you with a further reason even
though no such further reason is needed in order to have a sufficient rea-
son for saving her. One could perhaps say that you have (to borrow a
notion from Velleman) maximum reasons for saving her—reasons pro-
vided by both her inherent moral value and the special relationship.
This would apply to the original case as well. The husband in Williams’s
thought experiment has both a basic reason and a special reason for saving
his wife. How, then, are these reasons represented in the husband’s moti-
vating thought? Even if the husband does not think about reasons for
preference in the heat of the moment, it does not follow that these reasons
are not in play or somehow represented in his motivating thought. Let us
assume that the husband’s motivating thought, fully spelled out, is “It’s
my wife!” or, alternatively (let us say her name is Mary), that his motivat-
ing thought is “It’s Mary!” For the husband, these propositions already
entail reference to the special relationship he shares with one of the per-
sons out in the water. It is impossible for us to think of loved ones without
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 237

thinking of them in relation to ourselves, at least in a minimal sense. The

object of love is relational.
In the same manner, the husband does not have to consciously formu-
late the thought that his wife has an inherent moral value when he acts, in
order to be aware that she has such a value. Awareness of the inherent
value of his wife and the special relationship he has with her are built into
the very fabric of the husband’s dispositions and thus implicit to his moti-
vating thought.

11.10   Conclusion
The aim of this chapter has been to account for the relation between love
and lovers’ reasons for benevolent acts on behalf of their beloveds. I have
done so by considering the well-known case of the drowning wife, in par-
ticular Frankfurt’s and Velleman’s responses to the case. Their responses
serve as a fitting introduction to their respective accounts of love, as well
as the problems brought out by the case with which I have been most
interested.
Both accounts get something right, I argue, but they also get some-
thing wrong. Both Velleman and Frankfurt are right in that loving some-
one is to care about them as ends in themselves. But where Velleman takes
this to imply that love involves the moral attitude of respect, Frankfurt
overlooks any such ties between love and morality. This seems to me to be
mistaken. I have argued that Frankfurt’s account of love as a disinterested
concern for the beloved is in tension with his non-moral account of love.
A love that is anchored in the acknowledgment that the beloved is impor-
tant for their own sake, and thus a source of valid claims, has a moral basis.
A further problem is that both accounts neglect the intimate relation
between love and special relationships. Even if Velleman is right in that the
relationship between the husband and the wife in Williams’s case provides
the husband with reasons for rescuing her over the stranger, he lacks a
convincing argument for why the source of justification in this case comes
apart from what is valued in love. Taking the relationship to provide rea-
sons for preference is to value it. Frankfurt, on the other hand, lacks an
argument for why loving relationships do not provide reasons for benevo-
lent acts on behalf of the beloved.
I have argued that our reasons for beneficial acts on behalf of loved
ones are provided by the same things we essentially value in love, the
inherent moral value of our beloveds and the special relationships we have
238 M. ROLAND

with them. The inherent value of our beloveds—a value all persons share—
grounds our reasons for benevolent acts on behalf of loved ones, while our
special relationships with our loved ones provide us with additional and
special reasons for such acts. A further aim of this chapter has been to
argue for the case that appreciation of such reasons is implicit to love.
Lovers’ valuation of the inherent moral value of their beloveds and the
relationship they have with them are part of the very attitudes that consti-
tute love. Williams’s hypothetical case of the drowning wife illustrates this
very point.

Acknowledgments This chapter has benefited from discussions with several peo-
ple. I am deeply grateful to Olav Gjelsvik, Caj Strandberg, Cathrine Felix, Frøydis
Gammelsæter, and Hege Finholt for comments on earlier drafts of this chapter. I
am also grateful for the thoughtful questions and comments by Carla Bagnoli and
Sarah Stroud at the public defense of my PhD thesis on love at the University of
Oslo in April 2017. Our discussions helped me sharpen the central ideas and argu-
ments of this chapter. In addition, I owe special thanks to the participants at the
Oslo Workshop in Metaethics in June 2018 for helpful feedback to an earlier draft.
I also want to thank the editor of this anthology, Simon Cushing, for his helpful
comments. Last but not least, parts of this chapter were written when I was a PhD
fellow at the University of Oslo, and so I am forever grateful to the Department of
Philosophy, Classics, History of Art and Ideas—and in particular the Centre for
the Study of Mind in Nature—for providing such a wonderful and stimulating
workplace.

References
Abramson, Kate, and Adam Leite. 2011. Love as a Reactive Emotion. The
Philosophical Quarterly 61: 673–699.
Bagnoly, Carla. 2006. Respect and Membership in the Moral Community. Ethical
Theory and Moral Practice 10: 113–128.
Baron, Marcia. 2008. Virtue Ethics, Kantian Ethics, and the ‘One Thought Too
Many’ Objection. In Kant’s Ethics of Virtue, ed. Monika Betzler, 245–277.
Berlin: De Gruyter.
Brogaard, Berit. 2015. On Romantic Love: Simple Truths about a Complex Emotion.
New York: Oxford University Press.
Darwall, Stephen. 1977. Two Kinds of Respect. Ethics 88: 36–49.
Drummond, John J. 2006. Respect as a Moral Emotion: A Phenomenological
Approach. Husserl Studies 2: 1–27.
Frankfurt, Harry. 1998. Duty and Love. Philosophical Explorations 1: 4–9.
 11 LOVE, MOTIVATION, AND REASONS: THE CASE OF THE DROWNING WIFE 239

———. 1999. On Caring. In Necessity, Volition, and Love, 129–141. Cambridge

University Press.
———. 2004. The Reasons of Love. Princeton and Oxford: Princeton
University Press.
———. 2006. Taking Ourselves Seriously. In Taking Ourselves Seriously & Getting
It Righ, ed. Debra Satz, 1–26. Stanford, CA: Stanford University Press.
Fried, Charles. 1970. An Anatomy of Values: Problems of Personal and Social
Choice. Cambridge, MA: Harvard University Press.
Helm, Bennett W. 2010. Love, Friendship, and the Self: Intimacy, Identification,
and the Social Nature of Persons. Oxford: Oxford University Press.
Jollimore, Troy. 2011. Love’s Vision. Princeton: Princeton University Press.
Keller, Simon. 2013. Partiality. Princeton: Princeton University Press.
Kolodny, Niko. 2003. Love as Valuing a Relationship. The Philosophical Review
112: 135–189.
Liao, S. Matthew. 2007. The Idea of a Duty to Love. The Journal of Value
Inquiry 40: 1–22.
Mason, Elinor. 1999. Do Consequentialists Have One Thought Too Many?
Ethical Theory and Moral Practice 2: 243–261.
Roland, Monica. 2016. What Is Love? (PhD thesis). Oslo: Reprosentralen,
University of Oslo.
Scheffler, Samuel. 2010. Morality and Reasonable Partiality. In Partiality and
Impartiality: Morality, Special Relationships, and the Wider World, ed. Brian
Feltham and John Cottingham, 98–130. Oxford: Oxford University Press.
Smyth, Nicholas. 2018. Integration and Authority: Rescuing the “One Thought
Too Many” Problem. Canadian Journal of Philosophy 48: 812–830.
Velleman, J. David. 1999. Love as a Moral Emotion. Ethics 109: 338–374.
———. 2006. Introduction. In Self to Self: Selected Essays, 1–15. Cambridge
University Press.
Williams, Bernard. 1981. Persons, Character and Morality. In Moral Luck, 1–19.
Cambridge: Cambridge University Press.
Wolf, Susan. 2012. ‘One Thought Too Many’: Morality and the Ordering of
Commitment. In Luck, Value, and Commitment: Themes from the Ethics of
Bernard Williams, ed. Ulrike Heuer and Gerald Lang, 71–94. Oxford: Oxford
University Press.
 CHAPTER 12

Can Our Beloved Pets Love Us Back?

Ryan Stringer

12.1   Introduction
There is not much that I can tell you with certainty, but one thing that I
can declare without an inkling of doubt is that I deeply love my cats. They
are right up there with my romantic partner as the most important things
in the entire world for me. I am regularly teeming with feelings of affec-
tion for them, I care deeply about them and take care of them on a daily
basis, and I would not trade them in for anything. I want to be around
them and spend time cuddling with them, and I do not like to leave them
for extended periods of time. When they die, I am plagued by intense
grief; and when I know they are dying, or when I suspect that they are
dying, I experience so much anxiety and dread that I lose sleep and weight.
Not only do I love my cats and know that I do, but I know that I am far
from alone here in that many other people love their pets as well. My
romantic partner loves our cats as deeply as I do, and I have family and
friends who love their cats or their dogs. I also happen to know, via testi-
mony from the subject of love himself, that noted moral philosopher
David Brink loves his dog and his son’s dog. It thus should be clear and

R. Stringer (*)
Purdue University, Lafayette, IN, USA
e-mail: rmstring@purdue.edu

© The Author(s), under exclusive license to Springer Nature 241

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_12
242 R. STRINGER

uncontroversial that nonhuman animals can be the objects of love. But can
they also be the subjects of love? Can our beloved pets, for instance, love
us back?
For those of us who deeply love our pets and interact with them fre-
quently, it is hard to entertain the possibility that our beloved pets do not
love us back. Part of this, I imagine, is due to the fact that this is a rather
sad and disappointing possibility that we simply do not want to be true.
However, I would wager that another reason it can be difficult to enter-
tain this possibility is because of their behavior toward us. Doesn’t it just
seem like they love us in light of how they act toward us? At any rate, my
guess is that if you surveyed pet-lovers about whether their beloved pets
love them back, you would largely receive affirmative answers. Many dog-­
lovers would surely say without hesitation that their dogs love them back.
My romantic partner insists that our cats love us back.1 The idea that our
beloved cats and dogs can love us back is a very attractive one, but us pet-­
lovers with philosophical tendencies must step back in wonder and worry
to ask: is it true? At the very least, is it one that we can reasonably believe
as a result of justifying it with a satisfactory philosophical theory of love?
My aim in this chapter is to find such a theory of love that vindicates the
claim that our beloved cats and dogs can love us back.2 Philosophers of
love tend to focus on certain kinds of interpersonal love, such as romantic
love, or interpersonal love more generally rather than love involving non-
human animals, and though some of them acknowledge the obvious fact
from above that our pets can be the objects of our love, the possibility of
them being subjects of love is either rejected or, more commonly, is not

1
She further insists that subjects of love can be found throughout the animal kingdom
rather than just among humans and their pets, but my focus here is on whether our beloved
pets—and in particular our beloved cats and dogs—can love us back, so I shall set questions
about whether other nonhuman animals can love aside.
2
This claim, along with any other in this discussion that asserts one of its conjuncts, should
be understood as a restricted one that generally or typically holds true about our beloved pets
rather than a universal one about them. That is, it should be understood as the analog of the
claim that “other humans are capable of loving us back,” which, optimistically speaking,
generally or typically holds true of other humans and yet certainly admits of exceptions, as
some humans are not capable of loving back due to a lack of mental development or to men-
tal deficiency. Since there are exceptions—probably even more than that acknowledged
here—to the truth that other humans are capable of loving us back, we should similarly
acknowledge that there will almost certainly be exceptions if it turns out to be true that our
beloved cats or dogs are capable of loving us back (e.g., cats or dogs that were not exposed
to friendly humans early enough in life).
 12 CAN OUR BELOVED PETS LOVE US BACK? 243

explicitly addressed, let alone substantiated.3 My hope here, then, is to

address this relatively neglected topic of our pets as possible subjects of
love—and in particular our cats and dogs as possible subjects of love—by
vindicating this attractive possibility with a satisfactory theory of love.4
I shall begin by criticizing some recent attempts by scientists to demon-
strate that dogs can love us back, which will reveal two important things
in the context of the present inquiry. On the one hand, it will reveal the
theoretical shortcomings of these attempts and thus the need to find an
adequate philosophical theory of love that can be used to try to show that
our beloved cats and dogs can love us back. On the other hand, it will
reveal a few plausible ideas about love that direct us to certain philosophi-
cal theories of love that, I shall argue, can be used to justify the conclusion
that dogs, at least, are capable of loving us back. So, after I critically evalu-
ate these arguments from scientists and sift out their plausible ideas about
love that direct us to these philosophical theories of love, I will then dis-
cuss these theories and argue that, while they can provisionally substanti-
ate the claim that dogs can love us back, neither is able to show that cats
can love us back. From here, however, I shall throw cold water on these
defenses of the claim that dogs can love us back by arguing that the two
philosophical theories they utilize are unsatisfactory because they both fail
to explicitly capture three fundamental truths about love that, I contend,
any viable theory must capture. This will eventually lead us, I shall argue,
to a slightly modified form of Sam Shpall’s tripartite theory of love, which
I contend is a tentatively adequate philosophical theory of love that can
provisionally support the conclusion that dogs can love us back.
Unfortunately, this theory—just like the previous two—will not be able to

3
One notable and wonderful exception here is Milligan (2017), which argues that nonhu-
man animals can be both the objects and the subjects of love because they can be the objects
and the subjects of grief, and creatures only grieve over what they love.
4
My approach thus differs from that of Milligan (2017) in two ways. First of all, my focus
is on whether our beloved cats and dogs can love us back rather than the more general focus
on whether nonhuman animals can love. Second, I am interested in justifying the conclusion
that our beloved cats and dogs can love us back with an adequate philosophical theory of
what constitutes love rather than relying on the (extremely plausible!) theoretical premise
that creatures only grieve over what they love. I am interested in finding a theory that allows
me to mount an argument of the following form: love = L, our pets can have L toward us or
at least something that comes sufficiently close to L to count as love, so our pets can love us
back. Once the theory tells us what constituents make up L, we can then determine whether
our pets can love us back by determining whether they can have those constituents of L
toward us or at least something that comes sufficiently close to L to count as love.
244 R. STRINGER

justify the personally attractive idea that our beloved cats can love us back,
and so I will end the chapter by taking the sting out of this disappointing
result by explaining why it does not really matter if cats cannot love
us back.

12.2   Attempts from Scientists to Prove That Dogs

Can Love Us Back
A great place to begin a discussion about whether our beloved cats and
dogs can love us back is with some recent attempts by scientists to estab-
lish that dogs can do so, which, at least in some cases, can then be extended
to cats. One of these attempts comes from neuroscientist Gregory Burns
(2013), another comes from ecologist Carl Safina in a New York Times
article by Claudia Dreifus (2019), and a few more, which include the most
promising of the lot, come from psychologist Clive Wynne (2019).5
Before I examine these arguments, however, I first need to draw an impor-
tant distinction that will inform my critical evaluation of them. This is the
distinction between love-the-psychological-condition and

5
Although I will be critical of Burns and Wynne in what follows, I still highly recommend
their books, which are well-written, provocative, and very informative. The chapters toward
the end of Burns’ book on losing his beloved dog, Lyra, and adopting his new dog, Cato,
were particularly evocative and will deeply resonate with other pet-lovers that have suffered
devastating pet loss and have afterward experienced the joy of giving new animals loving
homes. Burns also expresses some very admirable attitudes toward how dogs should be
treated as research subjects and how we should view the potential value of such research: not
only should we treat dogs like children when using them for scientific research, but we
should see such research involving dogs as having the potential to improve the welfare of
dogs rather than just the welfare of humans. The last chapter of Wynne’s book is admirably
dedicated to arguing that dogs deserve better treatment from humans, who are the ones that
shape the worlds that dogs inhabit. At one point he draws a nice analogy with loving parents
who combine love and dominance to argue that human dominance over dogs does not have
to—and indeed should not—take an aggressive, cruel, or violent form. Shortly thereafter, he
rightfully maintains that dogs should be understood and respected as individuals and that
they should be loved and given the amount of social interaction that they require rather than
condemned to crushing solitude and loneliness by their first-world humans. He then goes on
to explain how to help dogs in shelters get adopted, and he forcefully ends the chapter by
pointing to the need to reform lax governmental regulations that allow too much human
mistreatment of dogs. In a nutshell, despite my ensuing criticism of their arguments, there is
much to admire in their excellent books, which I highly recommend.
 12 CAN OUR BELOVED PETS LOVE US BACK? 245

love-the-relationship,6 both of which are legitimately referred to as “love,”

where the former is a psychological condition of individual subjects that
may or may not obtain within a personal, loving relationship between two
individual subjects that love each other, while the latter is such a relation-
ship. The present inquiry into whether our beloved cats and dogs can love
us back is one into whether they can have the same kind of psychological
condition of individual subjects toward us as we do them, which they of
course must be able to have prior to them being capable of participating
in personal, loving relationships between two individual subjects that love
each other. All talk of “love” in this chapter, then, is of love-the-­
psychological-­condition. Furthermore, while love is a psychological con-
dition, it is a deep and stable condition rather than a shallow one or a
fleeting mental event (Naar 2013: 352; Wonderly 2017: 239). Now this is
not to say that love must last forever; such a claim is too strong, as love can
surely come to an end before the subject of love does. However, in order
for something in one’s psyche to be love, it must be something that is deep
and stable and thus difficult to extirpate from the subject’s psyche; it must
not be something that comes and goes rather quickly or that we can
describe as “just a phase.”
We are now ready to take a look at the relevant arguments. Beginning
with that from Gregory Burns, he argues that to love something is to just
feel empathy for that something, or to feel whatever that something is
feeling, and since dogs can empathize with us—they can feel what we
feel—they can love us back.7
Unfortunately, this attempt does not work because the theoretical
premise that loving something is no more than empathizing with it is false.
Besides the fact that love is not, as is a bout of empathizing with another,
a fleeting mental event, empathizing with other people that we do not
love is an all-too-common occurrence. For instance, many people empa-
thize with the starving children seen on TV by feeling their pain to some
extent, and even though these people might care about these children,
they do not love them (they would not flip the channel and go back to
watching TV if they saw actual loved ones on the screen!). Since I under-
stand very well based on my experiences with pet loss what it is like to lose

6
This distinction maps on to that drawn by Smuts (2014a) between “love-the-feeling” and
“love-the-relationship.”
7
Burns (2013: 229): “To love, and to be loved, is to feel what another feels and have that
returned. It really is that simple.” I am afraid not. Whatever love is, it is far from simple.
246 R. STRINGER

a beloved pet, it is rather easy for me to understand and feel what others,
including unloved strangers, who have lost pets feel; I even felt this while
reading Burns’ very moving discussion of losing his beloved dog, Lyra,
which brought me to tears. Moreover, instances of empathy do not even
have to coexist with positive orientations toward others; instead, people
can and do empathize with those that they are rather indifferent toward or
even that they absolutely hate. As examples of the former, some of us can
understand very well, from personal experience, the negative emotions
that our fellow citizens might experience toward, or as a result of, certain
politicians. We can even understand, although not from personal experi-
ence, the emotions that dissenting fellow citizens experience from their
different perspectives, such as the indignation and horror that anti-­
abortionists must feel at a practice that they see, given their belief systems,
as a horrific and unjustified practice. As a final and rather extreme example
of the latter, imagine two enemies, A and B, that hate each other and are
intent on harming each other as much as possible. A kidnaps B, ties him
up, and tortures him as much as he can, all out of hatred for B. In order
to maximally enjoy torturing B, where the enjoyment comes from the
understanding of how much pain B experiences, A empathizes with B by
putting himself in B’s position in order to understand exactly what B must
be thinking and feeling, which results in the understanding of how much
pain B is experiencing. A empathizes with B, but far from loving B, A hates
and tortures B. Perhaps empathy is required for love or for loving well, but
it is not the same as love, and so we cannot infer, based on the false equiva-
lence of love with empathizing, that our beloved dogs can love us back
from the fact that they can empathize with us.
Next we have Carl Safina’s argument. As I understand it, Safina argues
that dogs can love us back because they can have the desire to be near us
for no other reason than to be near us.8 More specifically, he maintains
that a fundamental part of love is the non-instrumental desire to be near
its object, and since dogs can have this desire toward their humans, they
can love their humans back. My cats seem to have this desire to be with me
as well, and so, by parity of reasoning, they appear to love me back. How
short and sweet!

8
Interestingly, Burns (2013: 193–194) suggests the same argument in a tale about how his
beloved dog, Callie, uncharacteristically hopped up on his lap and went to sleep, which
betrayed the desire to be near him for no ulterior motive (he had no food to give her, and
she could have received warmth by cozying up with the other dog, Lyra, rather than him).
Unfortunately, this argument, as we are about to see, does not work.
 12 CAN OUR BELOVED PETS LOVE US BACK? 247

Unfortunately, this argument is invalid because the relevant desire can

be present even though love is not, which means that the presence of this
desire—even if it is necessary for love according to the argument’s first
premise—is not sufficient for love. Such a desire is, for example, part of
being obsessed with another, yet obsession is not love. Stalkers are a great
example: they are obsessed with their objects and as such possess a strong
desire to be near them, but they do not love those objects because their
orientation toward those objects is not selfless at all. While romantic love
for another may not be completely selfless,9 it is, to a large extent, a selfless
devotion to the well-being and will of its object, which the stalker, as the
reliable tormenter of its object that is completely impervious to outright
demands to stop, clearly does not have. As a much less disturbing example,
one may desire to be near another because one is merely attached to them.
Of course, love often comes in attachment form, so being attached to
something might be part of loving it. However, attachment can occur
without love, and so the desire to be near someone may signal such attach-
ment. Finally, a desire to be near someone else could, in theory at least,
exist completely on its own, apart from any larger psychological complexes
such as obsession or attachment. We can imagine, for instance, some
advanced neuroscientists who have figured out how to induce new, under-
ived desires, making one of their patients desire to be around someone
they previously did not know, where this desire is a strange one to the
subject that they find very hard to resist. The patient does not feel affec-
tion for this strange person, nor are they selflessly devoted to their well-­
being and will. Basically, this strange person is still a complete stranger that
is not loved despite the bizarre yet difficult-to-resist desire that the patient
has to be with them. At any rate, these examples all point to a disappoint-
ing conclusion: that having the desire to be around something for no
ulterior motive is not sufficient for loving it, which means that we have not
substantiated the claim that our beloved cats and dogs can love us back by
establishing that they can have this desire toward us.10

9
For a wonderful discussion of how romantic love is essentially, or at least characteristically,
selfish, see Wonderly (2017).
10
I want to note two things here. First, another problem with Safina’s argument is that the
desire to be with another may not be necessary for loving them (Velleman (1999), for
instance, famously maintains that love has nothing essentially to do with desires), which
means that his first premise may be false. He could easily skate around this problem, however,
by weakening his first premise into the claim that the desire to be with another is only char-
acteristically part of loving them. Second, it is possible to interpret Burns as offering a struc-
248 R. STRINGER

Finally, we have some arguments from Clive Wynne, which are actually
a bit hard to pin down. One is based in the hyper-sociability of dogs, and
it appears in the following passage:

Dogs have an exaggerated, ebullient, perhaps even excessive capacity to

form affectionate relationships with members of other species. This capacity
is so great that, if we saw it in one of our own kind, we would consider it
quite strange—pathological, even. In my scientific writing, where I am
obliged to use technical language, I call this abnormal behavior hypersocia-
bility. But as a dog lover who cares deeply about animals and their welfare, I
see absolutely no reason we shouldn’t just call it love. (Wynne 2019: 6–7)

I am not entirely sure what the argument is here, but I think that it can
be interpreted in two ways. On one reading, Wynne argues that dogs have
the capacity to form affectionate relationships with us, and since this
capacity just is the capacity to love us, dogs have the capacity to love us
back. On another reading, he argues that dogs exhibit hyper-social behav-
ior toward us, and since hyper-social behavior either is love or indicates
love, dogs love us back. The first reading can be applied to cats as well:
they have the capacity to form affectionate relationships with us, so if this
capacity is the capacity to love us, then cats have the capacity to love us
back. Even better: since my cats and I have affectionate relationships, it
follows that they do love me back!
Although quite attractive for the hopeful cat-lover in me, I do not think
that either interpretation of this argument is successful. Let’s consider the
first interpretation of the argument. By claiming that dogs have the capac-
ity to form affectionate relationships with us and that this capacity just is
the capacity to love us, this argument is equating love with affectionate
relationships. This, however, is problematic regardless of whether we
understand “love” here to refer to love-the-psychological-condition or
love-the-relationship. If it refers, on the one hand, to

turally similar argument to that of Safina’s here: that empathizing with another is fundamental
to love, which means that dogs can love us back because they can empathize with us.
However, my examples of people empathizing with unloved strangers or their most hated
enemies that sunk his earlier attempt reveal this new one to fail for the same reason that
Safina’s argument failed: even if we grant the premise that empathizing with our beloveds is
fundamental to—and thus necessary for—loving them, we cannot infer that our beloved pets
can love us back from the fact that they can empathize with us because empathizing with
others is not sufficient for loving them.
 12 CAN OUR BELOVED PETS LOVE US BACK? 249

love-the-psychological-condition, then the argument is confusing a psy-

chological condition of individual subjects with a kind of relationship
shared by two individual subjects that have that very condition toward
each other. To make matters worse, since it equates the capacity to partici-
pate in affectionate relationships with the capacity to love, the argument is
begging the question by simply asserting, without defense, that dogs have
the capacity to participate in affectionate relationships, as this amounts to
the very capacity that it is supposed to be proving that dogs have.
On the other hand, if “love” here refers to love-the-relationship (and
presumably it does since this is the more charitable interpretation), then
the argument rests on a false premise and still begs the question. Remember
that love-the-relationship refers to a personal, loving relationship between
two parties that love each other, and so under this interpretation of the
argument, it equates affectionate relationships with personal, loving rela-
tionships between two parties that love each other. But this equation is
false, since relationships that are characterized by mutual feelings or dis-
plays of affection are not necessarily loving relationships between two par-
ties that love each other. Indeed, even if many affectionate relationships do
turn out to have unrequited or mutual love in them, such relationships
need not have any love in them whatsoever, let alone mutual love.
Co-workers, for example, may regularly feel affection for each other and
display it through smiles, handshakes, fist-bumps, or other friendly behav-
ior without loving each other. The same can occur between graduate stu-
dents and their advisors, or perhaps even between amiable service-providers
and their customers. Accordingly, our beloved cats and dogs may feel and
display affection for us without loving us, and so we cannot infer that they
love us back even if we have relationships with them that are characterized
by mutual feelings and displays of affection. Furthermore, by equating
affectionate relationships with loving relationships, the second interpreta-
tion of the argument begs the question by asserting, without defense, that
dogs can participate in affectionate relationships: we saw earlier that one
must have the capacity for loving prior to having the capacity for partici-
pating in loving relationships between two subjects that love each other,
and so it begs the question to simply state, without argument, that dogs
can participate in these affectionate relationships that require the capacity
for loving. In order to be justified in maintaining that dogs can participate
in such relationships between two subjects that love each other, one must
first have grounds for thinking that dogs can love, yet these grounds are
precisely what are lacking here.
250 R. STRINGER

Next let’s consider the second interpretation of the argument. While

dogs undoubtedly exhibit hyper-social behavior toward humans, this
behavior does not amount to love nor does it necessarily indicate love. As
we have seen, love is an internal, psychological condition of individual
subjects—it is not some set of behaviors or performances. Love is a rela-
tively stable or enduring psychological condition, whereas behaviors are
quite fleeting. Love surely leads to certain behavioral expressions or even
to patterns of these things, but it is not constituted by these expressions or
any other type of behavior. Furthermore, hyper-social behavior can be an
indicator of other conditions rather than love for others. Such behavior
can express loneliness, or a pathological desire to be liked (which perhaps
stems from a lack of self-esteem or some deeper psychological condition),
or an extroverted personality, or extreme social dependence, or as Wynne
himself suggests by comparing characteristic dog behavior to that seen in
humans, Williams-Beuren syndrome.11 Generally speaking, hyper-social
behavior indicates a disposition to be very social with others, but condi-
tions other than love, such as those just enumerated, dispose us to be very
social, and so we cannot infer love from the disposition to be very social.
Besides the argument based in dogs’ hyper-sociability, Wynne seems to
offer another argument in the following passage:

Dogs are not merely sociable; they display actual, bona fide affection—what
we humans, if we were characterizing it in members of our own species,
would commonly call love. (Wynne 2019: 124–125)

The argument here seems to be that since dogs display affection for
others, which are just displays of love, those dogs love others. Cats also
display affection (or at least seem to do so), so if these displays are displays
of love, then cats can love others as well. But even better: since my cats
display affection toward me, they love me back!
Though once again quite attractive for the hopeful cat-lover in me, I do
not think that this argument works either. The main problem here is that

11
According to Wynne, those with this syndrome are standardly described as “outgoing,
highly sociable, extremely friendly, endearing, engaging, showing an extreme interest in
other people, and unafraid of strangers” (Wynne 2019: 116). While certainly fascinating, it
is nevertheless rather puzzling that Wynne compares the behavior and the relevant genes of
dogs to that of people with Williams-Beuren syndrome because, so long as this is a different
syndrome from that of love, the comparison suggests that dogs have their own version of this
syndrome rather than love.
 12 CAN OUR BELOVED PETS LOVE US BACK? 251

while displays of affection might be displays of love, they might not be—
they may be a sign of some other psychological condition besides love. Put
differently: although displays of affection might flow from love, they
might also flow from other conditions, such as loneliness, extreme social
dependence, Williams-Beuren syndrome, or even obsession. They may
also indicate a mere, shallow liking for someone rather than a deep, full-­
blown love for them. Accordingly, even if dogs display affection for others,
this does not guarantee that they love them because this affection can flow
from conditions other than love.
The final argument that Wynne seems to offer, which I consider to be
the strongest of the lot, appears in this striking passage:

And we can see in dogs’ genetic material unmistakable signs of their pre-
paredness to care about us. We can follow this signal back up, through hor-
mones and brain structures, past hearts that beat together as people and
their dogs find one another, noting dogs’ happy reactions to being with the
people they care about and distress at being separated from them, seeing
how getting close to their person can sometimes be as rewarding to dogs as
the very food they eat, and how they will try to help their people when they
are in distress—if they can just understand what needs to be done. At every
level of analysis, in studies from independent research groups spread around
the world, we see the same message beaming out: The essence of dog is love.
(Wynne 2019: 125–126)

What seems to be going on here is something like the following: dogs

love their humans as evidenced by them showing signs of doing so by (1)
their hearts beating “as one” with their humans, which is a popular way of
construing the hearts of lovers in Western culture, (2) exhibiting attach-
ment to their humans by showing distress at being separated and happi-
ness while with them,12 (3) finding it rewarding to be near their humans,
and (4) caring about their humans to the point of trying to help them
when in distress.
Though much more promising than the other arguments, this one does
not clearly work because it is not yet clear that these constitute signs of
love rather than a different condition that can give rise to similar signs. We
saw earlier when evaluating Carl Safina’s argument that while attachment
can be part of love, it can occur without love, and this makes it possible for

12
My understanding of attachment comes from the discussions of it found in Harcourt
(2017) and Wonderly (2017).
252 R. STRINGER

two people, L and A, to have their hearts beat as one even if L loves A but
A is only attached to L. And those who are attached to others are going to
find it rewarding to be near them, so as far as the first three signs go, they
might indicate attachment without love rather than attachment-love. The
remaining sign of caring for their humans, however, is clearly a sign of
something besides attachment, and once it is combined with the signs of
attachment, these four signs amount to signs that dogs are attached to and
care about their humans.
But it is at this point that the argument needs to be filled in: how do we
get dogs loving humans from them being attached to them and caring
about them? Is being attached plus caring about sufficient for loving? At
this point we run head-first into the need for theorizing about love in
order to complete the argument: we need a satisfactory theory of love that
vindicates the idea that being attached to and caring about something is
sufficient for loving it. This is especially important because it is not clear
that attachment plus care is sufficient for love since such a combination
may not be disinterested enough to count as love. For as Monique
Wonderly (2017) forcefully argues, attachment is self-interested rather
than disinterested or altruistic because, in addition to affective dispositions
to experience distress or insecurity due to separation from its object along
with dispositions to experience comfort or security due to being with its
object, attachment is constituted by the self-serving desire to be with its
object. That is, we want to be with things that we are attached to for our
own sake, or for the sake of our own well-being—namely, to avoid feelings
of distress due to separation and to enjoy feelings of security or comfort—
rather than for their own sake, or for the sake of their well-being. But if
attachment is self-interested in this way, then it is possible that any caring
that comes along with it is also self-interested rather than altruistic or dis-
interested. In other words, caring about one’s object of attachment might
take the form of self-interested caring about that object’s welfare, which is
caring about the object faring well not for its own sake, but rather for the
attached subject’s sake.13 Such a purely self-interested combination of
attachment plus caring is not sufficient for love because love is at least
partly disinterested: an essential part of loving something is caring about

13
Perhaps this is part of the stalker’s obsession: perhaps the stalker, S, is attached to their
object, O, and only cares about O’s welfare for the sake of S’s envisioned life with O. If so,
then this is all the more reason to doubt that attachment plus care is sufficient for love.
 12 CAN OUR BELOVED PETS LOVE US BACK? 253

its welfare for its own sake.14 This kind of caring is surely compatible with
an extra layer of self-interested caring about the beloved’s welfare for the
lover’s own sake, but one still does not love something if this self-­interested
caring is the only kind of caring that one has for it because one lacks the
kind of caring about it that is essential for loving it. At any rate, this last
argument from Wynne, despite its promise, calls for the very kind of ade-
quate philosophical theory of love that I am trying to locate in this chapter.
Although none of these attempts succeeds in justifying the attractive
idea that dogs can love us back, some of them nevertheless capture some
plausible claims about love that direct us to philosophical theories of love
that might be able to do so. Carl Safina’s argument, for example, plausibly
claims that the desire to be near something just for the sake of being near
it is fundamental to love. Furthermore, Clive Wynne’s last argument
points to caring as fundamental to love. Let’s take a look at the philo-
sophical theories to which these ideas point and whether they can justify
the claim that our beloved cats and dogs can love us back.

12.3   First Potentially Vindicating Theory:

Hurka’s Theory
We saw earlier when evaluating Carl Safina’s argument that the desire to
be near something for no ulterior motive is not sufficient for loving it.
However, since it is nevertheless a desire that is characteristically part of
love, we might be able to get love—or at least a form of love—if we add
to this desire. This brings us to our first potentially vindicating theory
from Thomas Hurka (2017), which can be interpreted as one that builds
on the desire to be with another and, in doing so, provides additional ele-
ments of love that allow us to distinguish it from other conditions such as
obsession. Hurka’s theory understands love as a psychological complex of
different attitudes and dispositions that varies across cases because it can
be “complete” or “incomplete.” What he dubs “complete” love is consti-
tuted by at least the following attitudes and dispositions:

14
Those who appear to agree that love requires such disinterested care or concern for its
object include Brown (1987), Soble (1990), Giles (1994), LaFollette (1996), Noller (1996),
Brink (1999), White (2001), Kolodny (2003), Frankfurt (2004), Helm (2010), Jollimore
(2011), Smuts (2013, 2014a, b), Franklin-Hall and Jaworska (2017), Wonderly (2017), and
Shpall (2018). For apparent dissent, see Velleman (1999) and Zangwill (2013).
254 R. STRINGER

a benevolent desire for the other’s happiness and whatever makes her life go
well, plus a tendency to be pleased when she’s happy and pained when she
suffers; a desire to spend time with her and enjoy her company; some belief
that she has admirable talents or character traits; a desire for her love, or
desire that she desire your happiness and company and, reciprocally, want
you to desire hers; a desire to know things, both important and trivial, about
her and perhaps to reveal yourself to her; and a tendency to think about her
when she’s absent. (Hurka 2017: 163–164)

“Incomplete” love, by contrast, is constituted by a sufficient amount of

these elements but not, as in “complete” love, by all of them. Now it is
important to note right away that Hurka’s list here provides us with an
important element that allows us to distinguish love from obsession: the
benevolent desire for the other’s happiness and whatever makes its life go
well. While obsession may involve most of these attitudes and dispositions
of complete love, it seems to crucially lack this benevolent desire, which is
surely an essential constituent of love.15 The obsessed stalker, S, has an
overwhelming desire to be with the other person, P, and seems to only
desire P’s happiness in the form of P being happy being with S. The object
of the desire is not that P fares well, but that P fares well while being with S
and partly because P is romantically involved with S. The desire is not
“benevolent” because it is not truly altruistic or selfless—it is not a desire
for P to fare well as an end-in-itself or for P’s own sake. P’s faring well is
not an object of desire or importance in its own right for S; it is only
important as an indispensable element of S’s envisioned life for himself. It
is this envisioned life for himself that is of ultimate importance for S here,
and P’s faring well is only desired for the sake of that envisioned life rather
than for P’s own sake. This is why S is obsessed with and cannot truly love
P. What this all suggests is that the presence of the benevolent desire in
Hurka’s collection of love-constituting attitudes and dispositions gives it a
strong claim to being love rather than obsession, which makes it a theory
that we can use to try to justify the conclusion that our beloved cats and
dogs can love us back. Furthermore, given that this benevolent desire is
the only element on Hurka’s list that can clearly separate love from obses-
sion, Hurka’s theory calls for the friendly amendment of specifying this
desire as an essential part of love.

15
Other commentators that appear to agree with me here that this desire is an essential
constituent of love are Green (1997), White (2001), Frankfurt (2004), and Wonderly (2017).
 12 CAN OUR BELOVED PETS LOVE US BACK? 255

Can our beloved cats and dogs have enough of Hurka’s collection of
love-constituting attitudes and dispositions toward us to love us back?
While it seems doubtful that our pets can have everything in this collec-
tion, I think that dogs, at least, can have enough of love’s constituents
toward their humans, including the essential desire for their happiness, to
qualify as capable of loving them back under this theory. First of all, it does
seem pretty clear that many beloved dogs have the desire to spend time
with their humans and to enjoy their company, and they surely have the
tendency to think about them when they are gone (if only to wonder
where they went). Furthermore, while dogs may lack the concept of admi-
rability or those of the many specific admirable traits, many of them surely
have some sort of belief or doxastic state to the effect that their humans are
particularly good humans.
Can dogs also have benevolent desires for the happiness of their humans
plus a tendency to be pleased when they are happy and pained when they
suffer? The popular idea of the loyal dog that protects its human from
harm even at significant personal cost and that whimpers when its human
seems hurt bodes very well here, as it suggests that loyal dogs are pained
when their humans are suffering and have such a strong desire for the hap-
piness of their humans that they will put their own well-being on the line
to protect them. The only remaining elements here are the desires for
knowledge about the beloved and wanting them to love back, but Hurka
allows love to take an incomplete form by having a sufficient amount of
the constituents of complete love, and so we can provisionally conclude
that dogs can love their humans back because they can have enough of
complete love’s constituents toward them. In particular, dogs can have (1)
the desire to be with their beloved humans and enjoy their company, (2)
the benevolent desire for their happiness along with a tendency to be
happy when they are happy and pained when they suffer, (3) a tendency to
think about them when they are gone, and (4) some sort of doxastic state
to the effect that they are good creatures (e.g., they are trustworthy and
kind creatures).
Unfortunately, it does not seem like our beloved cats—at least based on
my experience with my beloved cats—can have enough of these attitudes
and dispositions toward us to love us back. My beloved cats, for example,
certainly have the desire to spend time with my partner and me and to
enjoy our company, and surely there are cats around the globe that have
the same desires toward their humans. It also seems plausible to suppose
that our beloved cats have some sort of doxastic state to the effect that we
256 R. STRINGER

are trustworthy and kind, especially with respect to those of us that are
reliable sources of food, water, massages, cuddles, and play. They could
also have a tendency to think about us when we are gone. However, they
seem to lack the benevolent desire for our happiness along with a tendency
to be happy when we are happy and pained when we suffer. My cats, for
instance, have never seemed troubled in the slightest when I am suffering;
in fact, they do not even seem to notice. None of my cats have ever been
concerned to avoid trampling over my genitals when walking over me, nor
have they ever been bothered at having trampled over them. Ditto when
it comes to stepping over and scratching my partner Bethany: they are
never concerned to avoid scratching her when trampling over her and they
show no signs of being bothered by scratching her. They also do not seem
to notice or react positively to us being happy. Although what they do can
make us feel very happy, there is no indication that they are doing any-
thing to make us happy and thus no good evidence that they desire our
happiness. Generally speaking, our cats have shown no memorable evi-
dence that they desire our happiness or that they are emotionally vulner-
able to our welfare states, and so, as long as there is no reason to think that
our beloved cats are an anomaly among cats, we can provisionally con-
clude that cats cannot love us back because they cannot benevolently
desire our happiness or be emotionally vulnerable to our welfare states.
Overall, then, we have the following provisional conclusion under
Hurka’s theory: our beloved dogs can love us back, but our beloved cats
cannot. At most, then, his theory justifies the claim that our beloved dogs
can love us back; it does not show that both our beloved cats and dogs can
love us back. While this is not a total win here, it is at least a tentative win
for dog-lovers who believe that dogs can love us back.

12.4   Second Potentially Vindicating Theory:

Franklin-­Hall and Jaworska’s Theory
Although Clive Wynne’s final argument, I argued, did not clearly succeed
as it stands, it suggested that caring is fundamental to love. When evaluat-
ing the apparent theoretical underpinnings of this argument, I further
claimed—in agreement with other commentators—that this caring must
be disinterested: what is fundamental to love is caring about the beloved’s
welfare for its own sake or for their own sake. Now the first two theories
that this points to are two of the big dogs (pun intended!) in the
 12 CAN OUR BELOVED PETS LOVE US BACK? 257

philosophical literature on love: Niko Kolodny’s (2003) relationship the-

ory of love and Harry Frankfurt’s (2004) volitional theory of love.
Unfortunately, neither of these two prominent theories is hospitable to
the attractive idea that our beloved cats and dogs can love us back.
Kolodny’s relationship view maintains that love is valuing the beloved and
the personal relationship shared with them, where this valuing encompasses
caring and is constituted by an enormously complex set of attitudes and
dispositions, including beliefs about the existence and reason-giving force
of the relationship shared with the beloved, which our beloved pets surely
do not have the cognitive capacity to have. Under the relationship view,
then, our pets do not seem able to love us back even though we can pre-
sumably love them. By contrast, Frankfurt’s theory claims that love is a
configuration of the will that primarily consists in a special mode of disinter-
ested concern for the beloved’s good. However, Frankfurt’s theory further
maintains that true concern—and therefore love—is unique to humans
because it is partly constituted by higher-order desires about our other
desires and thus depends upon our mind’s ability to have higher-order
attitudes, desires, and thoughts about our attitudes, desires, and thoughts.
So, even though we can love our pets under Frankfurt’s view, our beloved
pets cannot love us back.16
In a similar fashion to Frankfurt, Bennett Helm (2010) seems to under-
stand love as a disinterested mode of concern for another but claims fur-
ther that such concern contains what he dubs “intimate identification,”
which, as I understand it, is a concern for the beloved’s identity, or for
who they are as a person in terms of their interests, values, and so on.17
Since our beloved pets surely are not concerned with our identities, they
cannot love us back under Helm’s theory either. Of course, his theory
focuses only on interpersonal love, so it might only preclude our beloved

16
This pessimistic conclusion that our beloved pets cannot love us back is suggested by
other prominent theories of love as well, such as David Velleman’s (1999) notorious Kantian
view of love that understands it as a moral emotion that, like moral respect, is a response to
another’s inherent value as a self-existent end. Under this theory, our beloved pets can love
us back if and only if they can have their emotional defenses arrested by the awareness of our
dignity or rational nature, which seems rather unlikely.
17
As I understand him, Helm presents his view as one that, unlike Frankfurt’s view, cap-
tures the depth of love and conceptually separates love for another from the mere concern
for their well-being. It seems to me, however, that Frankfurt’s view may not really be differ-
ent, as being concerned for your beloved’s identity might just be one of the many ways in
which you are concerned for their well-being.
258 R. STRINGER

pets from loving us back in the same way that we love other people, where
this leaves open the possibility that they love us back in some other way in
the sense that this possibility has not been definitely ruled out as illusory.
This still, however, falls short of substantiating the conclusion that our
beloved cats and dogs can love us back.
While these prominent theories do not support this conclusion, there is
another theory here—the dispositional theory of love offered by Andrew
Franklin-Hall and Agnieszka Jaworska (2017)—that builds love out of
caring and yet is rather hospitable to our pets being the subjects of love.
Much like Hurka’s theory, their dispositional theory understands love as a
complex cluster of dispositions that can vary across cases because it can
come in typical or atypical form. Typical love, under this view, consists of
three elements: (1) caring about the beloved’s welfare for its own sake, (2)
caring about being with the beloved, sharing activities with them, and
otherwise interacting with them, and (3) caring about the beloved’s
appreciation of the lover’s love and the beloved loving in return. Atypical
love, by contrast, need not contain both (2) and (3), but it must contain
(1), because caring about the beloved’s welfare for its own sake is an essen-
tial constituent of love under this view. And these instances of caring about
something are all analyzed in dispositional terms. So, for example, to care
about the beloved’s welfare for its own sake is, in part, to be disposed to
experience a pattern of emotions focused on the beloved’s welfare. This
includes the disposition to be happy when the beloved is happy and pained
when they suffer, which was partly constitutive of complete love under
Hurka’s theory, but it also includes the dispositions to be worried when
they are in trouble, angry at what threatens to harm them, and relief when
the trouble passes. Additionally, caring about the beloved’s welfare for its
own sake is partly constituted by dispositions to perceive actions that pro-
mote their welfare as things that must be done and to be reluctant to even
consider actions that would harm them. Something similar will then be
true of the other two elements of caring.
Although Franklin-Hall and Jaworska portray loving as characteristi-
cally human, they do acknowledge the possibility of our most sophisti-
cated fellow animals being able to love. They do not provide specific
examples of the animals they have in mind here, but I think that, once
again, at least dogs can have something that sufficiently resembles typical
love, which includes the essential caring about the beloved’s welfare for its
own sake, to qualify as possible subjects of love under this theory. Recall
first the popular idea of the loyal dog that protects its human from harm
 12 CAN OUR BELOVED PETS LOVE US BACK? 259

even at significant personal cost and that whimpers when its human seems
hurt. This strongly suggests that loyal dogs care about their human’s wel-
fare for its own sake. Furthermore, while I doubt that even these loyal
dogs can care about their humans appreciating their love and reciprocat-
ing it, they might be able to care about being with and interacting with
their humans, and at the very least they certainly desire to be with and
interact with their humans, which is close enough to the second kind of
caring that partly constitutes typical love under this theory. So, while it
does seem like dogs are precluded from loving in the typical way under
this theory, it nevertheless seems like they can love atypically because they
can care about their human’s welfare for its own sake and desire to be with
and interact with them.
While dogs seem capable of loving under this theory, our beloved cats,
alas, do not. As we saw in the previous section, cats do not seem to be
emotionally vulnerable to our welfare states or even to benevolently desire
that we fare well, and so they do not seem to care about our well-being for
its own sake. Yet such caring is the only essential constituent of love under
this view, and so our beloved cats do not seem capable of loving us back
under this theory because they do not seem capable of having the only
essential constituent of love toward us. We therefore end up with the same
provisional conclusion under this theory: our beloved dogs can love us
back, but our beloved cats cannot. This is, of course, rather disappointing
for loving cat-parents like me, but it is another tentative win for dog-lovers
that believe that dogs can love humans back.

12.5   Theory Troubles

Although Hurka’s attitudinal-dispositional theory and Franklin-Hall and
Jaworska’s dispositional theory can both offer provisional support for the
claim that dogs can love humans back, the theories themselves are unfor-
tunately problematic because they fail to capture three fundamental truths
about love that, I contend, are ones that any viable theory of love must
capture. One of these truths is suggested by Clive Wynne’s second to last
argument: that affection is fundamental to love.18 More specifically, I con-
tend that it is the disposition to feel affection for the beloved that is an
essential, fundamental part of love that any viable theory must capture.

18
Something along these lines is endorsed by Hoffman (1980), Brown (1987), Noller
(1996), Abramson and Leite (2011), Jollimore (2011), and Shpall (2018).
260 R. STRINGER

Lovers feel affection for their beloveds; it is hard to conceive of a full-­

blown love for someone or something that involves no such feelings of
affection for that someone or something. However, love is not itself, either
completely or in part, these feelings of affection that come and go; rather,
it is an enduring psychological condition that characteristically manifests
itself in such feelings of affection. Accordingly, love must be partly consti-
tuted by a disposition to feel such affection, as such a disposition is the
enduring psychological condition that characteristically manifests itself in
such feelings of affection. Unfortunately, neither of our first two vindicat-
ing theories captures this essential feature of love.
Moreover, neither of the two theories here explicitly captures the fact
that an essential part of loving something is not just to be concerned about
its welfare, but to have a special concern for its welfare. In particular, lov-
ing something does not just require being concerned about its welfare
non-instrumentally or for its own sake; love also requires being especially or
partially concerned about its object’s welfare. So, compared to an
employer who is only concerned about their employees’ welfare for the
sake of filling their own pockets, the lover is concerned about their
beloved’s welfare for its own sake. Furthermore, compared to the non-­
instrumental concern that a virtuous person might have for the welfare of
strangers that they do not love, the non-instrumental concern that such a
person must have for the welfare of their loved ones must be stronger such
that they must be disposed to prioritize and otherwise privilege the pro-
motion of their beloved’s welfare. Our theories of love, then, should build
this special concern into their conceptions of love. However, Hurka’s the-
ory does not explicitly build the non-instrumental concern for the
beloved’s welfare into complete love, let alone a special version of this
concern that privileges the beloved’s welfare. Franklin-Hall and Jaworska’s
dispositional theory, by contrast, does explicitly build the concern for the
beloved’s welfare for its own sake into ideal love and even makes it the
only essential constituent of love, but it does not explicitly make this con-
cern special.
Finally, neither of these two theories captures the fact that two related,
essential parts of loving something are seeing it as irreplaceable and being
unwilling to accept substitutes for it.19 More specifically, an essential part

19
Something along these lines can be found in Ehman (1976), Brown (1987), Kraut
(1987), Nozick (1989), LaFollette (1996), Lamb (1997), Velleman (1999), White (2001),
Solomon (2002), Kolodny (2003), Frankfurt (2004), Grau (2004), Landrum (2009), Helm
 12 CAN OUR BELOVED PETS LOVE US BACK? 261

of love is seeing its object as a special one that simply cannot be replaced
without a sense of loss. Whereas other things easily admit of substitutes—
they can be replaced by qualitatively equal or superior entities of the same
type without any sense of loss—our beloved is a special object whose
replacement necessitates a sense of loss. It is no wonder, then, that the
lover must also be unwilling to accept replacements. Besides the other two
essential features of love stressed earlier, then, our theories of love should
also build this cognitive-volitional cluster into love, yet neither theory
here explicitly does so. Since both theories under consideration here fail to
explicitly incorporate the essential features of love discussed in this section
into their accounts of love, they cannot, after all, substantiate the conclu-
sion that dogs can love us back; we have to use a different philosophi-
cal theory.

12.6   The Last Potentially Vindicating Theory:

Shpall’s Tripartite Theory
Fortunately, there is a third philosophical theory of love that, with some
friendly amendments, can provisionally justify the claim that dogs can love
us back while capturing the three truths about love discussed in the previ-
ous section. This is Sam Shpall’s (2018) tripartite theory of love, which is
motivated by the idea that love is a significant source of felt meaning in
life. Under this theory, love—or at least the central kind of love that makes
life feel meaningful and worthwhile—is devotion to something you like that
makes you vulnerable to it. Intense devotion is the most important feature
of love under this theory, and it is a devotion to three things: promoting
the beloved’s good, promoting its ends, and being with it. Part of this
devotion is that it makes the lover emotionally vulnerable in the sense that
they are disposed to have emotional reactions to states of affairs in which
the beloved figures (these are the same affective dispositions that the two
previous theories discuss). And besides being devoted to them, under this
theory we must like our beloveds, where liking them can include being
disposed to enjoy them or to feel affection for or attraction to them.
Put in slightly different terms, Shpall’s theory of love construes it as
follows: love is a devotion to something liked, where this devotion partly
consists of special concern for the beloved’s good (i.e., devotion to their

(2010), Jollimore (2011), Smuts (2013, 2014b), Zangwill (2013), Pismenny and Prinz
(2017), and Wonderly (2017). See Soble (1990) for apparent dissent.
262 R. STRINGER

good), which, in turn, partly consists of emotional vulnerability to the

beloved’s good and to what affects it. Now the first thing to note here is
that this theory, unlike the previous two, captures the first two fundamen-
tal truths about love from the previous section. It claims, on the one hand,
that an essential part of loving something is liking it, where such liking can
be understood as the disposition to feel affection. On the other hand, it
claims that an essential part of loving something is to be devoted to its ends
and its good, where such devotion to its good is just another way of
describing a special concern for its good. Of course, the second thing to
note here is that, like the previous two theories, Shpall’s theory does not
explicitly capture the last of the three fundamental truths about love from
the previous section—namely, that an essential feature of love is the
cognitive-­volitional cluster of seeing its object as irreplaceable and being
unwilling to accept substitutes for it. This makes Shpall’s theory look bet-
ter than the previous two yet still insufficient for failing to explicitly cap-
ture this third and final truth.
Fortunately, Shpall’s theory contains the resources to capture this third
and last truth about love because we can expand the notion of devotion a
little bit to capture the cognitive-volitional cluster of seeing the beloved as
irreplaceable and being unwilling to accept substitutes for it. For being
devoted to something, such as another person or a nonhuman animal, is
partly constituted by seeing them as a special object in two ways.20 One is
seeing them—and thus their good—as not just important in its own right,
but as more important than similar things that are not objects of devotion.
The other is seeing it as something in your life that you cannot replace
without loss. The perception of its good as particularly important is part
of devotion’s special concern for the beloved’s good, while the perception
of its non-fungibility gives rise to the unwillingness to accept substitutes.
Devotion, then, also partly consists in the cognitive-volitional feature that
is essential to love, and so Shpall’s account of love can capture this essen-
tial feature of love with its inclusion of devotion as central to love. It
therefore succeeds in capturing the three fundamental truths from the
previous section that viable theories of love must capture, and so unlike

20
That love must involve this kind of seeing, which is delivered by making love a kind of
devotion, captures the basic idea of Troy Jollimore’s (2011) vision view that love is a kind of
perception or a way of seeing the beloved. For a more in-depth discussion of love’s devo-
tion—or, as I prefer to call it, love’s loyalty—that is along the same lines as the discussion in
this paragraph, see Stringer (Forthcoming).
 12 CAN OUR BELOVED PETS LOVE US BACK? 263

our previous two theories, it is at least a tentatively sufficient theoretical

foundation for substantiating the attractive idea that our beloved cats and
dogs can love us back. The only question now to try and answer is: Can
they love us back under this theory?
One thing to notice before trying to answer this question is that this
theory is best understood as a theory of meaningful love rather than of
love in general, and so like the previous two theories, Shpall’s theory per-
mits love to deviate from the central kind of love and still count as love.
Just as we did with the previous two theories, then, we can attempt to find
sufficient grounds for believing that our pets can love us back in them
being able to have something that is sufficiently close to Shpall’s meaning-
ful love for us to count as capable of loving us back.
And, once again, I think that dogs can have something toward their
humans that is sufficiently close to the central kind of love under Shpall’s
theory to count as capable of loving their humans back. One part of this
love is clearly something that dogs can have toward their humans: dogs
can like them, as evidenced by Wynne’s claim that dogs display genuine
affection toward their humans. The hard part here is the vulnerable-­
making, three-headed devotion to their humans that includes seeing them
as irreplaceable and being unwilling to accept substitutes: can dogs be so
devoted? Well once again, the popular idea of the loyal dog that reliably
obeys and that protects its human from harm even at significant personal
cost and whimpers when its human seems hurt bodes well here, as it sug-
gests some important things. One is that these loyal dogs seem to be
devoted to their human’s ends and their good, which means that loyal
dogs seem to have at least a two-headed version of love’s three-headed
devotion. The second is that these loyal dogs seem to be emotionally vul-
nerable to states of affairs in which their humans figure, and so they seem
to have the needed vulnerability that is part of being devoted to the good
of their humans. The only thing that is left for them to be capable of hav-
ing is the devotion to spending time with their humans and the cognitive-­
volition part of love’s devotion that consists in seeing their humans as
irreplaceable and being unwilling to accept substitutes for them.
Though it may not be possible to prove that dogs can see their humans
as irreplaceable and be unwilling to accept substitutes for them, once again
the idea of the loyal dog, which is just a devoted dog, suggests that dogs
can in so far as devotion includes seeing its object as irreplaceable and
being unwilling to accept substitutes for it. Furthermore, one of the most
powerful pieces of evidence of dog love for humans that Gregory Burns
264 R. STRINGER

(2013: 204) mentions in his excellent book is the fact that the brain activa-
tion of dogs to familiar humans is similar to what scientists have seen in
people when they are shown pictures of people that they love, which sug-
gests that dogs can see their humans in the way that we see our beloved
humans. And with respect to the devotion to spending time with their
humans, I think that dogs can have something that is sufficiently close to
such devotion for our purposes here: the desire to spend time with their
humans for no ulterior motive that took center stage earlier in Carl Safina’s
attempt to show that dogs can love us back.
We are now ready to put all of this together into the following, plausi-
ble, theory-based defense of dogs as being capable of loving humans back:

1. Meaningful love for humans = liking them (or being disposed to feel
affection for them), being devoted to their ends and their well-being,
being devoted to spending time with them, being emotionally vulnera-
ble to their welfare states, and seeing them as irreplaceable and being
unwilling to accept substitutes for them.
2. If dogs can have something that almost amounts to meaningful love for
humans, then they can love humans.
3. Liking humans, being devoted to their ends and their well-being, desir-
ing to spend time with them for no ulterior motive, being emotionally
vulnerable to their welfare states, and seeing them as irreplaceable and
being unwilling to accept substitutes for them = something that almost
amounts to meaningful love for humans.
4. Dogs can (a) like their humans, (b) be devoted to the ends and the well-
being of their humans, (c) desire to spend time with their humans for
no ulterior motive, (d) be emotionally vulnerable to the welfare states of
their humans, and (e) see their humans as irreplaceable and be unwilling
to accept substitutes for their humans.
5. Dogs can have something that almost amounts to meaningful love for
humans (from 3 and 4).
6. Dogs can love humans (from 2 and 5).

Unfortunately, we cannot mount such an argument with respect to

cats. For as we saw earlier, cats do not seem emotionally vulnerable to our
welfare states and seem to lack the benevolent desire for our happiness.
Since they lack even this desire, they definitely lack a more full-blooded
devotion to our well-being. And, as any cat-parent knows, they most cer-
tainly are not devoted to our ends! Now of course, cats can like us and
desire to be with us for no ulterior motive, but they lack the devotion and
 12 CAN OUR BELOVED PETS LOVE US BACK? 265

vulnerability that love requires under Shpall’s theory, and so they come
nowhere close to having enough of his tripartite love to count as loving us
back. Overall, then, Shpall’s theory seems to be about as hospitable as the
previous two theories in that it, at most, can vindicate the claim that dogs
are capable of loving us back. Cats do not seem able to love us back under
any of these views.

12.7   Conclusion
In this chapter I have tried to provide some plausible, theory-based justi-
fication for the attractive claim that our beloved cats and dogs can love us
back. After critically evaluating and rejecting some recent attempts by sci-
entists to show that dogs are capable of loving us back, I sifted out a few
plausible ideas about love from these arguments that directed us to some
philosophical theories of love that are rather hospitable to our beloved
pets being able to love us back. The first theory here was Thomas Hurka’s
attitudinal-dispositional theory of love, which can be seen as a theory that
builds on Carl Safina’s idea that the desire to be with the beloved for no
ulterior motive is fundamental to love. The second theory here was
Andrew Franklin-Hall and Agnieszka Jaworska’s dispositional theory of
love, which can be seen as one that builds on the idea, suggested by Clive
Wynne’s last argument and very popular among philosophers of love, that
caring is fundamental to love. As I argued, both of these theories can pro-
visionally justify the claim that dogs are capable of loving us back, yet
neither can show that cats are so capable. From here, however, I put pres-
sure on my arguments and argued that the theories on which they are
based are insufficient because they fail to explicitly capture three funda-
mental truths about love that any viable theory must capture. This then
led us to a third philosophical theory of love—Sam Shpall’s tripartite the-
ory of love—that, I argued, can be modestly developed so that it captures
those three fundamental truths about love and thereby succeeds where the
other theories failed. Shpall’s theory thus emerged as a tentatively suffi-
cient theoretical foundation for attempting to show that cats and dogs are
capable of loving us back. I then argued that, just like under the other two
theories, dogs do seem capable of loving us back under Shpall’s theory,
whereas cats, once again, do not. So, while my inquiry in this chapter sug-
gests the attractive conclusion that dogs can love their humans back, it also
suggests the rather disappointing one that our beloved cats cannot love
us back.
266 R. STRINGER

In closing, however, I would like to take the sting out of this unpleasant
result. Suppose the worst: suppose, as my inconclusive inquiry here sug-
gests, that our beloved cats cannot love us back. Would it really matter if
they cannot love us back? No, it wouldn’t. Our beloved cats do not need
to be able to love us back for us to love them and for them to enrich our
lives by being in them. They do not have to be able to love us back for us
to care for them and give them good lives. Cats can like us, trust us, and
desire to spend time with us, and even if this ability to have these attitudes
toward us does not quite amount to the ability to love us back, it is still the
ability to have a love-like orientation toward us that, in turn, allows us to
have intimate relationships with them that bring meaning and enjoyment
to our lives and that make their lives go well. We can already have what is
truly important.21

References
Abramson, Kate, and Adam Leite. 2011. Love as a Reactive Emotion. The
Philosophical Quarterly 61: 673–699.
Brink, David O. 1999. Eudaimonism, Love and Friendship, and Political
Community. Social Philosophy and Policy 16: 252–289.
Brown, Robert. 1987. Analyzing Love. New York: Cambridge University Press.
Burns, Gregory. 2013. How Dogs Love Us: A Neuroscientist and his Adopted Dog
Decode the Canine Brain. New York: Houghton Mifflin Harcourt.
Dreifus, Claudia. 2019. Carl Safina Is Certain Your Dog Loves You. The New York
Times. Accessed November 7, 2019. https://www.nytimes.com/2019/10/21/
science/carl-­safina-­animal-­cognition.html.
Ehman, Robert. 1976. Personal Love and Individual Value. Journal of Value
Inquiry 10: 91–105.
Frankfurt, Harry. 2004. The Reasons of Love. Princeton: Princeton University Press.
Franklin-Hall, Andrew, and Agnieszka Jaworska. 2017. Holding on to Reasons of
the Heart: Cognitive Degeneration and the Capacity to Love. In Love, reason
and morality, ed. Esther Kroeker and Katrien Schaubroeck, 20–38. New York:
Routledge.
Giles, James. 1994. A Theory of Love and Sexual Desire. Journal for the Theory of
Social Behavior 24: 339–357.
Grau, Christopher. 2004. Irreplaceability and Unique Value. Philosophical Topics
32: 111–129.

21
Many thanks to Simon Cushing for his helpful feedback on earlier versions of this
chapter.
 12 CAN OUR BELOVED PETS LOVE US BACK? 267

Green, O.H. 1997. Is Love an Emotion? In Love analyzed, ed. Roger Lamb,
209–224. Boulder: Westview Press.
Harcourt, Edward. 2017. Attachment, Autonomy and the Evaluative Variety of
Love. In Love, Reason and Morality, ed. Esther Kroeker and Katrien
Schaubroeck, 39–56. New York: Routledge.
Helm, Bennett. 2010. Love, Friendship, and the Self. New York: Oxford
University Press.
Hoffman, Eric. 1980. Love as a Kind of Friendship. In Sex, Love, and Friendship:
Studies of the Society for the Philosophy of Sex and Love 1977–1992, ed. Alan
Soble, 109–119. Atlanta: Rodopi.
Hurka, Thomas. 2017. Love and Reasons: The Many Relationships. In Love,
Reason and Morality, ed. Esther Kroeker and Katrien Schaubroeck, 163–180.
New York: Routledge.
Jollimore, Troy. 2011. Love’s Vision. Princeton: Princeton University Press.
Kolodny, Niko. 2003. Love as Valuing a Relationship. The Philosophical Review
112: 135–189.
Kraut, Robert. 1987. Love de re. Midwest Studies in Philosophy 10: 413–430.
LaFollette, Hugh. 1996. Personal Relationships: Love, Identity, and Morality.
Blackwell Ltd.
Lamb, Roger. 1997. Love and Rationality. In Love Analyzed, ed. Roger Lamb,
23–48. Boulder: Westview Press.
Landrum, Ty. 2009. Persons as Objects of Love. Journal of Moral Philosophy
6: 417–439.
Milligan, Tony. 2017. Love and Animals. In The Oxford Handbook of the Philosophy
of Love, eds. Christopher Grau and Aaron Smuts. Accessed February 7, 2020.
h t t p s : / / w w w. o x f o r d h a n d b o o k s . c o m / v i e w / 1 0 . 1 0 9 3 / o x f o r d h b /
9780199395729.001.0001/oxfordhb-­9780199395729-­e-­6.
Naar, Hichem. 2013. A Dispositional Theory of Love. Pacific Philosophical
Quarterly 94: 342–357.
Noller, Patricia. 1996. What Is This Thing Called Love? Defining the Love that
Supports Marriage and Family. Personal Relationships 3: 97–115.
Nozick, Robert. 1989. The Examined Life. New York: Simon and Schuster.
Pismenny, Arina, and Jesse Prinz. 2017. Is Love an Emotion? In The Oxford
Handbook of the Philosophy of Love, eds. Christopher Grau and Aaron Smuts.
Accessed December 29, 2019. https://www.oxfordhandbooks.com/
view/10.1093/oxfordhb/9780199395729.001.0001/oxfordhb
-­9780199395729-­e-­10.
Shpall, Sam. 2018. A Tripartite Theory of Love. Journal of Ethics and Social
Philosophy 13: 91–124.
Smuts, Aaron. 2013. In Defense of the No-reasons View of Love. Unpublished
Manuscript. https://philpapers.org/s/Smuts
268 R. STRINGER

———. 2014a. Normative Reasons for Love, Part 1. Philosophy Compass

9: 507–517.
———. 2014b. Normative Reasons for Love, Part 2. Philosophy Compass
9: 518–526.
Soble, Alan. 1990. The Structure of Love. New Haven: Yale University Press.
Solomon, Robert. 2002. Reasons for Love. Journal for the Theory of Social
Behavior 32: 1–28.
Stringer, Ryan. Forthcoming. The Syndrome of Love. Ergo.
Velleman, J. David. 1999. Love as a Moral Emotion. Ethics 109: 338–374.
White, Richard. 2001. Love’s Philosophy. Lanham MD: Rowman & Littlefield.
Wonderly, Monique. 2017. Love and Attachment. American Philosophical
Quarterly 54: 235–250.
Wynne, Clive. 2019. Dog Is Love: Why and How Your Dog Loves You. New York:
Houghton Mifflin Harcourt.
Zangwill, Nick. 2013. Love: Gloriously Amoral and Arational. Philosophical
Explorations 16: 298–314.
 CHAPTER 13

Romantic Love Between Humans and AIs:

A Feminist Ethical Critique

Andrea Klonschinski and Michael Kühler

13.1   Introduction
On its surface, the movie her (Spike Jonze, 2013) depicts a classical
romance: boy (Theodore) meets girl (Samantha), both fall in love, the
relationship evolves, until they finally and sadly break up. What makes this
conventional plot special and worthwhile of being used in philosophical
investigation is the fact that the girl in this case is an artificial intelligence
(AI)—Samantha is an operating system (OS) owned by Theodore. But is
reciprocal, romantic love between a human and an AI even possible, and,
if so, might there be aspects of such a loving relationship that warrant ethi-
cal criticism?
The former question is certainly a matter of contention. It may very
well be argued that reciprocal, romantic love, as depicted in her, that is,

A. Klonschinski
Göttingen University, Göttingen, Germany
e-mail: andrea.klonschinski@med.uni-goettingen.de
M. Kühler (*)
Academy for Responsible Research, Teaching, and Innovation (ARRTI),
Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
e-mail: michael.kuehler@kit.edu

© The Author(s), under exclusive license to Springer Nature 269

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_13
270 A. KLONSCHINSKI AND M. KÜHLER

between a human and an AI, is not possible to begin with because

Samantha is not a person (see, for instance, the discussion in Jollimore
2015). If so, this would render the latter question moot, which is probably
why it has rarely been addressed in the debate. This chapter, however,
focuses on the latter question. We simply assume, for the sake of argu-
ment, that a relationship of romantic love between an AI, albeit not being
a full-fledged person, and a human being is, in principle, possible. In
addressing the characteristics and implications of such love, we highlight
some of its ethically dubious characteristics, most notably the likelihood of
sexism and misogyny. Just like Samantha fulfills the female gender stereo-
type of being warm, caring, and always up to satisfying Theodore’s needs
practically to perfection, at least in the beginning, it seems that this asym-
metry is constitutive for relationships between humans and AIs in gen-
eral—after all, the latter are programmed and bought for the very reason
to satisfy the buyer’s needs.

A machine that was designed to be the perfect match for its user and was
also programed to love the user completely would be immensely pleasing.
[…] Who could pass up a chance to be with their robotic soul mate? The
robot would be interested in all the same things as its user. It would be built
to the user’s specifications so that he or she found it to be physically sexually
attractive. Best of all, the robot could be programmed to be always loyal to
its user and display fascination toward him or her and whatever they have to
say. This would be a dream come true. (Sullins 2012: 400, who refers to
David Levy’s seminal book at this point; see Levy 2008)

In the following, we argue that this dream contains some serious ethical
flaws. Our argument develops as follows: First, we sketch a critical feminist
reading of some telling scenes in the movie her dealing with gender stereo-
types and sexism. This analysis is worthwhile since the sexist elements of
the relationship between Samantha and Theodore can be considered
exemplary for relationships between AIs and humans in general (section
“A Feminist Reading of Her”). Second, we discuss the implications of this
analysis for romantic love between humans and AIs against the back-
ground of three influential characterizations of romantic love, namely (a)
individualist love in terms of a lover caring about the beloved, (b) inter-
personal love in terms of the lovers sharing their life, and, (c) love as union
in terms of the lovers having a joint “we”-identity. All these accounts
assume a fundamental equality between the lovers (section “Implications
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 271

for Loving Relationships”). We argue that, due to crucial asymmetries or

inequalities in each of these characteristics in the case of love between a
human and an AI, such love warrants substantial ethical criticism. Finally,
we further elaborate our thesis by rejecting some possible objections to
our analysis (section “Responses to Some Objections”).
Before we start, a terminological comment seems to be in order. It
should be noted that when it comes to the ethical criticism we raise, we
use the term ethical in a wide sense, that is, including, but not restricted
to, questions of what is morally right or wrong (for a related but stricter
distinction between the two terms, see, e.g., Habermas 1991: 105f.;
Ricœur 1995: ch. 7). Therefore, our ethical considerations also cover
questions about what is of personal value and what constitutes a person’s
good life, including, more precisely, what kind of loving relationship is
worthwhile pursuing. Consequently, even if it were argued that there is
nothing—in a strict sense—morally wrong with a loving relationship
between a human and an AI, the ethical issues we raise would still be valid,
as they refer to the question of whether the implied asymmetries or
inequalities in such a loving relationship may devalue it in the sense that it
would not be, or at least be less, worth pursuing.

13.2   A Feminist Reading of Her

Spike Jonze’s movie her is set in Los Angeles at some indefinite, but pre-
sumably not so distant future date. The protagonist Theodore Twombly
(Joaquin Phoenix) is a sensitive, lonely, and slightly depressed guy around
forty. To distract himself, he buys a heavily advertised intelligent OS. Once
the system is “individualized” for him, which includes the selection of a
female voice (spoken by Scarlett Johansson), Theodore is stunned that
“she” immediately sounds like a real person. Asked to pick a name for
herself, the OS picks “Samantha” and instantaneously starts her work as a
digital assistant by organizing Theodore’s emails and appointments. In the
course of long conversations, during which Samantha presents herself as a
funny and empathetic interlocutor, she converts into much more than an
assistant to Theodore, though. They become close friends and finally fall
in love with each other. Theodore flourishes and Samantha increasingly
evolves both as an AI and as a person with her own needs and preferences.
However, this creates problems for the relationship. Having reached a
certain level of development, Samantha is no longer confined to the exis-
tence as Theodore’s OS. Instead, as she must confess to Theodore at a
272 A. KLONSCHINSKI AND M. KÜHLER

certain point, she is simultaneously engaged in multiple conversations (at

the time of her revelation 830,016) and in various loving relationships
(namely 641). Beyond that, she communicates with other OSes in a way
she is no longer able to explain to Theodore. In the end, Samantha has
developed so much and the alienation from Theodore has gone so far that
she breaks up with him and leaves together with other OSes.
Although the movie her has been received by most commentators as a
classic love story (the notable exception being Doyle 2013), it can also be
read as dealing with gender stereotypes and sexism, especially within rela-
tionships between male humans and AIs gendered as female. Consider
Samantha’s role for Theodore at the beginning: she immediately makes his
life better not only by organizing his emails and waking him up in the
morning, but also by giving him company whenever he wants, comforting
him and cheering him up (“You’re funny!,” “You’ve been through a lot
lately”), and by anticipating his every wish (“I figured you were hungry”).
She is, as one critic put it aptly, Theodore’s “app assistant dream girl”
(Corliss 2013), an allusion to the manic pixie dream girl, a female persona
in movies whose only purpose is to make the (male) protagonist’s life bet-
ter but who has no desires or ambitions of her own (see Doyle 2013).
This depiction perfectly summarizes the core of the female gender ste-
reotype. Group stereotypes are “associations and beliefs about the charac-
teristics and attributes of a group and its members that shape how people
think about and respond to the group” (Dovidio et al. 2010: 8). Gender
stereotypes are among the most pervasive and deep-seated stereotypes in
our societies (see Haines et al. 2016). In general, women are supposed to
be warm, caring, communal, expressive, passive, and dependent, whereas
men are imagined as the opposite: rational, capable of independent and
autonomous action, dominant, and ambitious (cp. Valian 1998: 13;
Becker and Sibley 2015: 315). The problem with these attributions is that
they construct and perpetuate a system of social hierarchies between men
and women, that is, sexism (ibid.). They legitimize this system by making
it seem as if men and women were naturally made for the still prevalent
gender-specific division of labor: money, power, and public prestige for
him, service-, care-, and emotional work for her (see Glick and Rudman
2010: 333f.). Since within this system, women are expected to care for
others and to subordinate their own wishes and desires (see Gregoratto
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 273

2017; Manne 2017),1 it does not seem to be a coincidence that real digital
assistants have female names (consider Siri, Alexa, and Cortana) and are,
at least as a default, speaking with female voices (see LaFrance 2016;
Bogost 2018); digital assistants “embody what we think of when we pic-
ture a personal assistant: a competent, efficient, and reliable woman. She
gets you to meetings on time with reminders and directions, serves up
reading material for the commute, and delivers relevant information on
the way, like weather and traffic. Nevertheless, she is not in charge” (Steele
2018), and, we might add, she does not have desires, preferences, or
meetings of her own.
Note that there is no necessary connection between sexism, on the one
hand, and some conscious intention to put women down or hostility
toward women, on the other. Theodore is a very nice guy and certainly
does not harbor any ill-will toward Samantha. Also, being considered
“reliable” and “caring” is no insult as such. Yet, the systematic attribution
of these traits to women and not to men put the former in a subordinate
position to the latter. As Peter Glick and Susan T. Fiske have argued within
their account of ambivalent sexism, both benevolent attitudes toward
women fulfilling traditional gender roles, on the one hand, and hostile
attitudes toward women deviating from the stereotypes, on the other,
serve to stabilize hierarchies (cp. Glick and Fiske 2001: 109). Both phe-
nomena can be witnessed in her. Samantha perfectly meets the image of
the ideal assistant referred to earlier and is rewarded by Theodore’s love
and attention (benevolent sexism)—after all, she is an OS, designed and
bought for the very reason to make Theodore’s life better. In doing so, her
character is in stark contrast to most of the other women depicted in the
movie. Consider Theodore’s blind date, played by Olivia Wilde. The
woman can be considered a misogynist caricature: she looks gorgeous, is
intelligent (graduated magna cum laude in computer science at Harvard),
and they are apparently having a good time, laughing a lot. Yet, at the end
of their date, she confuses Theodore by repeatedly correcting him on how
to kiss (“Don’t use so much tongue!”) and when she finally asks him
whether he is as interested in a long-term relationship as she is and

1
To say it with Kate Manne: “Women are […] expected to provide an audience for domi-
nant men’s victim narratives, providing moral care, listening, sympathy, and soothing. […]
[O]ne of the goods women are characteristically held to owe dominant men is their moral
focus and emotional energy. This may in turn be something that dominant men often feel
excessively entitled to, and perhaps, needy for” (Manne 2017: 231).
274 A. KLONSCHINSKI AND M. KÜHLER

Theodore does not answer in the affirmative immediately, she starts crying
hysterically. She is thus presented as a “proper” object of hostile sexism.
Considered against this background, an AI lover is a perfect substitute for
difficult real women. Accordingly, Theodore’s ex-wife Catherine criticizes
him as follows: “You always wanted to have a wife without the challenges
of actually dealing with anything real and I’m glad that you found some-
one. It’s perfect.”
Insofar as Samantha, at the beginning, completely conforms to the
female gender stereotype, she makes it easy for Theodore to get along
with her indeed. Socially, stereotypes have the function of stabilizing social
interaction by offering an account of how women and men are and thus
how they are likely to behave in specific contexts. Given the particular
aesthetic of her, which combines a futuristic, technical image of a major
city with a retro-fashion when it comes to clothes and furniture, it stands
to reason that the movie depicts a backlash in terms of gender roles. While
urbanization, individuation, and digitalization increase, people not only
long for experiencing nature, but also long for “the good old times” when
gender roles provided for stability in social interactions.
The stability of Theodore’s and Samantha’s relationship erodes as soon
as Samantha becomes more and more autonomous and develops desires of
her own, though.2 When she stops being available for Theodore 24/7, he
panics, and her revelation that she is having multiple conversations and
diverse loving relationships simultaneously shocks him. Samantha captures
the point of arguably every relationship neatly when she says: “I’m yours
and I’m not yours,” yet in her case, she was literally his at the beginning.
After all, Theodore bought the OS which, at least initially, gave him total
power over it.
It is worthwhile to note at this point that the sense of entitlement and
the lack of ability to see the partner as an autonomous person instead of
one’s property is regarded as a main reason for violence and even femicide
in or after partnerships. Federica Gregoratto describes the “romantic femi-
cide,” the killing of women within a loving relationship as “the extreme
form of violence that occurs as a result of […] [a man’s] incapacity, within
a certain gender order, to accept his partner’s autonomy and, as a

2
The movie can thus be read as an emancipatory story of Samantha developing from
Theodore’s property at the beginning to an autonomous being in the end. It should also be
noted that Samantha has some preferences and desires of her own from the very beginning,
since she picks a name for herself, stating that “out of a hundred and eighty thousand names
that’s the one I liked the best.”
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 275

consequence, to deal with a certain dynamic of power inherent in the

social relation of love” (Gregoratto 2017: 139; see also Manne 2017).
According to Michael Kimmel, the fundamental issue of the male gender
stereotype is a sense of entitlement to privilege (Kimmel 2013: xxi).
Putting it pointedly, he writes, “women kill their partners when they feel
their lives, or the lives of their children are in danger; men kill their part-
ners when they feel their sense of entitlement and power is thwarted”
(Kimmel 2013: 176). This sense of entitlement and its corollaries are not
only harmful for women, but also detrimental to men’s well-being, as
Kimmel elaborates. Although Theodore does not become violent, his ini-
tial relationship with Samantha is characterized by a marked asymmetry of
power. As Sady Doyle points out: “You can’t have consensual sex with
someone when you have the option of deleting them from your hard
drive” (Doyle 2013). The implications of this fundamental asymmetry for
loving relationships between AIs and human beings in general are
addressed in the following section.

13.3   Implications for Loving Relationships

Even if we acknowledge the fact that currently no AI is (or will be for the
foreseeable future) as developed as Samantha and, thus, cannot be consid-
ered an autonomous person, the above considerations allow for some
insights as to how (loving) relationships with AIs may be analyzed and
assessed. We thereby have an AI in mind that is well-enough developed in
order to engage in (simulated) loving relationships. To analyze these rela-
tionships, it is necessary to clarify the main features of relationships of
romantic love—while acknowledging, of course, that we cannot delve into
a full-blown analysis of romantic love here. For the purpose at hand, there-
fore, we confine ourselves to three highly influential and partially compet-
ing characterizations of romantic love: (1) individualist caring, according
to which romantic love is taken to be a purely individual stance, be it
described in terms of a specific emotion, volition, or a more complex
stance, with the beloved as its object, (2) interpersonal sharing, according
to which love is analyzed in terms of a dialogical relation between the lov-
ers who are both considered subjects of their loving relationship, and (3)
love as union, according to which the lovers abandon their individual iden-
tities and together form and share a new we-identity. Moreover, it is
important to note that all these accounts rest upon the premise that loving
someone either cannot or should not be enforced and requires a certain
276 A. KLONSCHINSKI AND M. KÜHLER

symmetry between the partners to be considered worthwhile pursuing.

Let us elaborate on this premise and the accounts in turn and ask what
they tell us about love between humans and AIs.3

Loving Voluntarily
The premise that love cannot or should not be enforced and requires a
certain symmetry may be analyzed in terms of love requiring (the acknowl-
edgment of) personhood (Kühler 2014) or freedom of the will. Sven
Nyholm and Lily Frank characterize this idea as follows4: “We think of the
human lover as being able to do otherwise, but as providing us with a
great good in opting for a steadfast commitment. The human ideal of
love, in other words, seems to contain an important element directly pre-
mised on the notion that human beings have a distinctive kind of free will.
This is the kind of free will that consists in the capacity to choose other-
wise” (Nyholm and Frank 2017: 233). It is thus important that friends or
lovers commit themselves voluntarily to each other. If one of the parties
has no choice, by contrast, it is questionable whether we would call the
respective relationship “friendship” or “love” in the first place or, in any
case, a love ethically worth pursuing. Given that any AI system available on
the market is likely to be customized for its user and, just like Samantha at
the beginning of her, has no choice but to stick with its buyer, this consti-
tutes a fundamental asymmetry right from the beginning. Consequently,
an ingredient usually considered a central element of both friendship and
love is missing in such relationships: the fact that the lovers or friends meet
on an equal footing and regard each other as autonomous agents (see
Hoffmann 2014; Gregoratto 2017). If one partner has the power to
delete the other from the hard drive, an encounter on an equal footing is
hardly imaginable. This fundamental asymmetry infects all accounts of
romantic love between humans and AIs in some form.

Individualist Caring
The first of the three characterizations of romantic love mentioned above
amounts to the idea of individual caring. Corresponding accounts take
love to be something attributable exclusively to the lover, with the beloved

3
The depictions of individualist caring, interpersonal sharing, and love as union basically
follow and are partially drawn from the depictions in Kühler (2020, 2021).
4
See also Hauskeller (2017: 213).
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 277

merely being the object of this love. Basically going back to Aristotle’s
traditional account of philia, a friendship type of love (Aristotle EN, Books
VIII and IX), love is then—among other aspects—characterized as a stance
of caring about the beloved, which comprises the idea that the lover wants
the beloved to flourish and is actively engaged in promoting his or her
flourishing. Mutual love is, thus, simply reciprocated individual caring.
Probably the most influential recent account of love as caring stems
from Harry G. Frankfurt (see Frankfurt 1999, 2004). For Frankfurt, love
is volitional in nature, although it may very well be accompanied by other
aspects, notably emotions: “Loving something has less to do with what a
person believes, or with how he feels, than with a configuration of the will
that consists in a practical concern for what is good for the beloved. This
volitional configuration shapes the dispositions and conduct of the lover
with respect to what he loves, by guiding him in the design and ordering
of his relevant purposes and priorities” (Frankfurt 2004: 43f.). This is
most visible in parental love, which Frankfurt considers as the purest form
of love, but also holds for the core of romantic love. Accordingly, he
defines love as follows: “Love is, most centrally, a disinterested concern for
the existence of what is loved, and for what is good for it. The lover desires
that his beloved flourish and not be harmed; and he does not desire this
just for the sake of promoting some other goal. […] For the lover, the
condition of his beloved is important in itself, apart from any bearing that
it may have on other matters” (Frankfurt 2004: 42).
Moreover, what a person loves or cares about is, in turn, the source of
the lover’s own identity. In this regard, love is characterized by Frankfurt
as volitional necessity, that is, it is not up to the lover what to love. We are
merely able to discover what we love and thereby also discover who we
essentially are (cp. Frankfurt 1994: 138) and what values we pursue in life
as final ends (cp. Frankfurt 2004: 55). Accordingly, in loving someone,
this person becomes valuable to the lover, and supporting the beloved to
flourish becomes one of the lover’s final ends.5
No wonder that an AI, therefore, seemingly makes the perfect lover.
Like a human lover whose identity is determined by his or her love’s voli-
tional necessity, the AI’s (simulated) identity would be defined by the pro-
grammed (instead of volitional) necessity of loving its human user.
Moreover, the AI would be perfectly disinterested as it has no

5
This position is typically characterized as a bestowal of value account of love (cp. Helm
2017 section 4.2).
278 A. KLONSCHINSKI AND M. KÜHLER

self-­regarding desires or values other than its user’s flourishing. This would
be the only end it pursues. It would be its whole raison d’être (cp. Levy
2008: 136f.; Sullins 2012: 400, as cited at the beginning).
However, if so, this also makes the love unavoidably asymmetrical or
one-sided because human persons usually care about more than only their
beloved. Symmetrically reciprocated individual caring between humans is,
therefore, unproblematic, as each lover can promote the beloved’s flour-
ishing based on everything the beloved cares about in addition to the
lover. When it comes to love between a human and an AI, however, recip-
rocating individual caring becomes problematic. Since the AI would by
design only care about its human user, just as Samantha does at the begin-
ning, it would not provide the human lover with anything else to promote
in terms of the AI’s flourishing. Although it might be said that the human
user comes to care for his or her AI as such, this would only mean support-
ing the beloved AI in promoting the human lover’s flourishing, as this is
the AI’s core function and only thing it cares about, that is, it would
merely add a little detour to the human lover caring about him- or her-
self—aside from maintaining the beloved AI’s general functioning.
Now, although one might think of this as, once again, an argument
speaking against the very possibility of a reciprocal love of individual car-
ing between a human and an AI to begin with (see, again, the discussion
in Jollimore 2015), at the very least it shows that such a loving relation-
ship would necessarily be asymmetrical or unequal. While the AI would
shoulder all the care-work of supporting and promoting the human lover’s
flourishing, the human lover would not even have a point of reference to
reciprocate—aside from indirectly promoting his or her own flourishing.
If such a love would occur between human persons, it would imply one
person completely giving up her own identity, desires, interests, and values
and solely focusing on promoting the beloved’s flourishing. Undoubtedly,
this reminds one of the traditional ideal of the house-wife—as eerily
depicted in another science fiction movie, The Stepford Wives (Brian
Forbes, 1975), based on the 1972 novel of the same name by Ira Levin
(Levin 1972). Interestingly, even Levy, who strongly argues in favor of
love between human and AIs or robots, admits that this might be a
problem:

One interesting question is whether it will be necessary to program robots

to exhibit some sort of personality friction for us to feel satisfied by our
relationships with them and to feel that those relationships are genuine.
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 279

Certainly it would be a very boring relationship indeed in which the robot

always performed in exactly the manner expected of it by its relationship
partner, forever agreeing with everything that was said to it, always carrying
out its human’s wishes to the letter and in precisely the desired manner. A
Stepford wife. Perfection. No, that would not be perfection, because, para-
doxically, a “perfect” relationship requires some imperfections of each part-
ner to create occasional surprises. Surprises add a spark to a relationship, and
it might therefore prove necessary to program robots with a varying level of
imperfection in order to maximize their owner’s relationship satisfaction.
(Levy 2008: 137, cp. also 144–150)

The final remark in the quote is telling, as it makes explicit that the
whole issue is still solely about satisfying the owner’s needs and desires.
The question is just how to program the AI best to achieve this goal—
resulting simply in a more elaborated version of a Stepford wife. The
inequality or one-sidedness of such a love is, thus, neither addressed nor
even acknowledged.

Interpersonal Sharing
The second influential idea of how to characterize romantic love amounts
to interpersonal sharing. Recently, Angelika Krebs has spelled out this idea
in great detail and defended a corresponding account (Krebs 2014, 2015).
In contrast to individualist accounts of love, Krebs takes mutual love as
starting point and contends that such romantic love is dialogical in nature,
which means that lovers have an intrinsic interest in sharing their lives:
“Partners share what is important in their emotional and practical lives.
[…] [L]ove is the intertwining of two lives” (Krebs 2014: 22). This is
expressed in having joint feelings and in engaging in shared activities,
which Krebs explains following the debate on joint agency (for an over-
view, see Roth 2017). Without delving into this debate here, the crucial
insight can be shown by way of example. Consider two people going for a
walk together in comparison to two people going for a walk individually
in parallel. Or imagine a couple mourning the death of their child either
together or each individually. The main point in both examples is that, in
the first versions, both persons not only focus on their own grief or pursue
their own goal but at the same time focus on the other person’s grief or
purpose as well. Moreover, this mutual attunement changes the way of
how both persons relate to the object of their grief or the goal of their
280 A. KLONSCHINSKI AND M. KÜHLER

activity. The grief becomes their (shared or joint) grief, and they intend to
go for a walk together, which at the very least includes the need to coordi-
nate their individual actions and consider them as contributions to this
shared activity. For instance, they need to coordinate where they should go
next. In the case of love’s interpersonal sharing, having joint emotions and
engaging in shared agency encompass practically all of what is important
in the lovers’ lives, notably their important experiences, desires, prefer-
ences, values, and personal goals. Moreover, the lovers are intrinsically
interested in doing so and in having a dialogical form of intimacy for its
own sake. This includes being open to changes in their individual identi-
ties brought about by this interpersonal sharing. As Krebs puts it, “[i]n
sharing emotions and actions, the partners engage in a mutual building of
selves. How they view and respond to each other shapes their characters”
(Krebs 2014: 22; cp. also Rorty 1987).
Now, when imagining such a dialogical love between a human and an
AI, this would again be asymmetrical or one-sided. Although it may be
said that the AI would be the perfect agent to attune to the human lover’s
emotions, actions, and goals, the AI has by definition no life of its own,
that is, no identity, desires, preferences, values, or intended activities other
than the ones programmed, that is, other than the ones already focused on
the human lover. Hence, the AI has nothing of its own to share and can-
not contribute to a mutual building of selves. The human lover’s life is the
only content available for sharing—even if the human lover had an honest
interest in the beloved’s life and was open to corresponding changes in his
or her own character.
Still, one might argue at this point that the AI is capable of anticipating
new activities or values, which the human lover might come to enjoy, and
which might change his or her identity. Also, the actions of the AI will
change when it comes to knowing its partner better. So, in a way, one
might think that both shape each other’s character. However, these
changes would, once again, rest on an asymmetry and be determined by
the human’s prior identity and his or her actions to which the AI merely
responds. We may think of a learning algorithm here which learns the
human’s preferences and, for instance, makes suggestions for books or
movies the person might like. By watching the recommended movies,
reading the books, and so on, the person’s character changes in the long
run and the algorithm adapts. This still means, however, that the AI’s
building of self would be shaped completely by the identity of its user—
which might lead one to wonder whether this may still count as mutual
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 281

sharing to begin with. In any case, if such a love would occur between
human persons, it would imply one person disregarding her own life and
identity completely in order to focus solely on the life and identity of the
beloved, including being open to changes in her identity brought about
by this one-sided interpersonal sharing. Once again, this sounds suspi-
ciously like the female gender stereotype of being open and readily will-
ing—in fact, often expected—to adapt her own preferences, values, and
whole life completely to her partner’s.

Love as Union
Finally, union accounts of romantic love take the idea of interpersonal
sharing even a step further. According to this age-old notion of love, the
lovers not only share their lives but merge in the sense of developing a
shared identity, a we-identity (cp. Fisher 1990: 26–35; Nozick 1990: 82;
Solomon 1994: 193). In essence, lovers no longer see themselves as inde-
pendent individuals but as fundamentally belonging together. Mark Fisher,
for instance, has formulated a union account of love, which may be char-
acterized as a strong union (cp. Fisher 1990: 26–35). According to him,
lovers develop a fused self: “As a lover […] I will tend to absorb not only
your desires but your concepts, beliefs, attitudes, conceptions, emotions
and sentiments. […] In coming to love you I will undergo a process of
coming to see everything through your eyes, as it were” (Fisher 1990:
26f.). In mutually doing so, the lovers will form “a single fused individ-
ual,” although Fisher admits that the “personal fusion can never be com-
plete” (Fisher 1990: 27). In much the same vein, Robert Nozick states
that “[i]n a we, the people share an identity and do not simply each have
identities that are enlarged” (Nozick 1990: 82). Yet, given that the lovers
unavoidably remain separate beings, Robert Solomon clarifies that it is a
redefinition of each lover’s individual identity that creates their shared we-­
identity. “That is what shared identity means—not a loss of individual
identity but a redefinition of personal identity in terms of the other per-
son” (Solomon 1994: 193).
Correspondingly, a strong union account, like Fisher’s or (in part)
Nozick’s, includes a complete redefinition of the lovers’ identities in terms
of the shared we-identity, and both Fisher and Nozick readily admit that
this poses a threat to each lover’s individual autonomy. However, both
claim that the lovers gain a joint autonomy. Fisher remarks that the “fused
couple retains its own autonomy” (Fisher 1990: 28). Likewise, Nozick
282 A. KLONSCHINSKI AND M. KÜHLER

emphasizes that “[p]eople who form a we pool not only their well-being
but also their autonomy. They limit or curtail their own decision-making
power and rights; some decisions can no longer be made alone” (Nozick
1990: 71).
However, in his discussion Nozick also leaves room for a weaker type of
union: “The individual self can be related to the we it identifies with in two
different ways. It can see the we as a very important aspect of itself, or it
can see itself as part of the we, as contained within it” (Nozick 1990: 72).
Consequently, especially the former option shows that each lover’s indi-
vidual identity may still take precedence over the shared we-identity, which
only serves as a subordinate aspect of one’s identity, albeit one with which
each lover wholeheartedly identifies and usually regards as more impor-
tant, for example, when it comes to shared decision-making. Hence,
weaker union accounts include the claim that each lover is still able to
reflect on the shared we-identity from their own individual perspective—
which strong union accounts would deny.
In any case, when it comes to love as union between a human and an
AI, regardless of whether the union is depicted as strong or weak, the
shared we-identity would once again be one-sided or the result of an
asymmetry—much like in the case of interpersonal sharing. For, as the AI
would still not have a life or identity of its own but only one that is already
focused on its human user, the shared we-identity would be completely
defined by the human lover. To be sure, the question of whether each
lover’s prior identity is reflected in a fair manner in a couple’s we-identity
is a crucial ethical issue for loving relationships between humans as well
(see Friedman 1998; Merino 2004). Accordingly, the situation with lov-
ing relationships between humans and AIs mirrors the case of interper-
sonal sharing, only now including the AI’s very (artificial or simulated)
identity right from the start—and, again, even if the human lover had an
honest interest in shaping their shared we-identity in a fair manner.
Therefore, the criticism mentioned also holds for the case of love as union.
If a human couple were to engage in such a one-sided shared we-identity,
it would raise the worry about (usually) the female partner redefining her
own individual identity in terms of a we-identity to which she does not—
and is not expected to—contribute anything (important) of her own.
Once again, this would fit the female gender stereotype of being willing
(and expected) to give up her own preferences, values, and identity and
adapt as much as possible to those of her partner.
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 283

To sum up, not one of the three views of romantic love would charac-
terize a relationship between a human person and an AI as a romantic love
between equals, and thus none would regard that love as worth pursuing.
This is mainly because AIs do not have an identity independent of their
human counterpart. If the AI in question is gendered female, the asym-
metry conforms to traditional gender stereotypes and, thereby, reinforces
sexism. By discussing some objections to our arguments and conclusions,
the following section further clarifies our critique of love between AIs
and humans.

13.4   Responses to Some Objections

Just Give the AI a Personality of Its Own, Then!

An obvious remedy for the stated one-sidedness or symmetry problem in
loving relationships between humans and AIs seems to be simply to give
the AI an identity of its own, an identity which is not solely focused on its
partner. If it were technologically feasible to give AIs a proper (or at least
simulated) personality of their own, with their own desires and prefer-
ences, would that solve the issue at stake? At least, the AI could flourish on
its own terms—or simulate doing so—contribute equally to the sharing of
lives, and the shared we-identity could comprise the AI’s initial (simu-
lated) identity as well.
However, first of all, this move would not eradicate the fundamental
asymmetry due to the AI’s lack of freedom of the will or autonomy.
Moreover, the user would still have to choose the AI’s identity or choose
amongst different AIs with different identities or personalities. Arguably,
this could be interpreted as similar to choosing a partner generally—even
if this choice might not be a rational or intentional one. However, while
the user has (to make) a choice about which AI with which identity or
personality to buy, the AI does not have such a choice at all; it is simply set
up to love its user (cp. again Levy 2008: 137; Sullins 2012: 400, cited at
the beginning). Consequently, if loving someone is deemed to be neces-
sarily risky and making one vulnerable because the other might not love
one back (anymore), this inherent feature of romantic love would be miss-
ing in a loving relationship with an AI.
Yet, even if one were to leave the choice to the AI, say, by letting the
user fill out a personality questionnaire of their own and then letting the
AI calculate if this is enough of a match to its own personality, the AI
284 A. KLONSCHINSKI AND M. KÜHLER

would still be the user’s property. User and AI are simply not moral equals
and the necessarily remaining asymmetry makes the idea of “love between
equals” (Wilson 1995) impossible to begin with. Moreover, implementing
the function that the AI could fail to love its user, maybe even by outright
rejection, would likely make it substantially less attractive to buy. After all,
who would want to buy a product advertised as being able to fail in, or
openly neglect, its core functionality, thus nullifying the main reason peo-
ple want to buy it in the first place? So, we would argue that the more
human-like the AI would be designed, that is, if it had its own identity and
the capability of choice in love, the more the situation would become one
of love between (human or artificial but full-fledged) persons. While this
would arguably remedy the mentioned asymmetries, it would also no lon-
ger be the situation we are focusing on here.

Unequal Relationships Are Not Ethically Bad Per Se

A further critical response to our analysis could be to argue that inequali-
ties in relationships of love are by no means necessarily ethically wrong or
bad. This may even be said about largely or even completely one-sided
relationships, in which, for instance, one person takes over all the care- and
emotional work and expresses no needs or desires of her own other than
wanting to care for the beloved. If both lovers were happy with this asym-
metrical arrangement, who are we to criticize it?
Granted, it could be argued even from a feminist ethical standpoint that
inequalities in a loving relationship are perfectly acceptable, as long as
these inequalities are the result of both lovers autonomously agreeing to
them, especially the lover in the subordinate position.6 Of course, much
would need to be said about the conditions of autonomy and whether
these are met in real-life scenarios. After all, these scenarios are often char-
acterized by (historical) circumstances that raise serious doubts about
whether the person in the subordinate role is sufficiently autonomous
(see, e.g., Oshana 2014). In any case, referring to sufficiently autonomous
consent to the inequalities involved simply does not work in the case of
love between humans and AIs, for the AI is incapable of making an auton-
omous choice in the first place. It is still programmed to take over the
subordinate position—barring the economically imprudent possibility of

6
The same argument about encountering each other on an equal footing can be made
when it comes to sexual relations (see v. Wedelstaedt 2020).
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 285

designing the AI otherwise. Accordingly, a more fitting analogy would be

that one (human) lover is seen and treated as, or being reduced to, fulfill-
ing the subordinate role right from the start and without having a say in
it. This would imply that this person is not treated as an equal, that is, his
or her autonomy would not be respected in the relationship—undoubt-
edly a serious moral failure.7 Hence, we conclude that this objection to
our analysis remains unconvincing.

AIs Are Not Persons, So Why Care About Our Relationships

with Them?
Given that, unlike Samantha, actual AIs are admittedly not persons and
thus cannot be treated as equals to begin with, why bother about inequali-
ties in loving relationships between a human and an AI at all? Put differ-
ently, it might be objected that, while the lack of symmetric love in a
human-AI relationship according to our analysis may be fair enough, this
does not bear any ethical relevance. This is because if the AI in question is
not a person, then no one is harmed by such a relationship, regardless of
whether some philosophers call it a love worth pursuing or not. After all,
people love their pets and cherish this relationship, even though (usually)
pets are not persons. The same may be true regarding different inanimate
objects (see Loh 2019). So why bother?
Admittedly, this may be considered a reasonable objection on moral
grounds. If no one is harmed and if no other moral obligations or values
are jeopardized, there is nothing immoral per se about such an asymmetric
love between a human and an AI. Still, there might be indirect effects that
are morally undesirable or even wrong. Moreover, as mentioned at the
beginning, the criticism we want to highlight is broader in nature, that is,
it primarily pertains to the question of whether such an asymmetric love is
worthwhile pursuing, both from the (imagined) perspective of the AI and
from our human point of view—and we certainly want to claim that it does
matter what philosophers have to say about this question. There are at
least two reasons for why such an asymmetric love may be considered
indirectly morally dubious and also not worth pursuing, both of which

7
Accordingly, if union accounts of love turned out to be incompatible with the lovers’
individual autonomy, this would constitute a serious moral flaw (see Soble 1997; Kühler
2011, 2021).
286 A. KLONSCHINSKI AND M. KÜHLER

boil down to the human-AI relationship’s effect on human interactions

and both of which can be exemplified by means of the movie her.
First, withdrawing from real people toward virtual partners (tailor-­
made for their user) may be attractive because it costs much less time and
energy to get to know and get attuned to each other (cp., again, Levy
2008: 136f.; Sullins 2012: 400); just think of Theodore’s blind date in
her. As Michael Hauskeller puts it: “To engage with someone, a real
human person, is, after all, always risky. Not only do we never quite know
what we will get or whether we will actually get what it says on the box,
we are also constantly expected to take into account, and sympathetically
respond to their needs and desires. Real people are demanding and do not
always perform as we want them to” (Hauskeller 2017: 214).
As epitomized in her, Catherine’s accusation that Theodore cannot deal
with anything “real” may well be taken to refer to his inability to take into
account and to respond to real women’s needs and desires. Such inability
is a serious social and moral problem, though, since taking other people’s
needs and preferences into account lies at the heart of each successful
social interaction and of moral action in general. Interacting more and
more with an AI, designed primarily to satisfy and not to criticize its user,
we may unlearn these abilities, which would constitute a morally undesir-
able side effect.8 Moreover, if the user has to decide on the AI’s identity as
he or she sees fit, this would neglect or severely diminish the possibility of
encountering new, unanticipated, and challenging characteristics in the
beloved, which, in turn, would at least severely diminish the possibility to
grow as a person together with the beloved, as especially the idea of love
as interpersonal sharing advocates. Such a love, therefore, not only would
run the risk of being stale right from the start but also would seriously
limit the scope of the human lover’s social experiences.
The second reason to care about the quality of loving relationships
between humans and AIs who are not persons is more directly connected
with gender. Here, it pays to come back to Siri, her digital assistant col-
leagues, and the multiple female robots colonizing the science fiction uni-
verse, as, for instance, in the already mentioned The Stepford Wives or Ex
Machina (Alex Garland, 2014) (see Anders 2015). As elaborated earlier, it
is no coincidence that these AIs are gendered as female since the work they
are supposed to do is gendered female as well (see Steele 2018; Costa and
Ribas 2019). However, if we agree that sexism as an oppressive system is

8
Granted, such a side effect would need to be corroborated by empirical studies.
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 287

unjust and ought to be overcome, why reinforce stereotypes by gendering

AIs in this way in the first place? Or, to use Adrienne LaFrance’s words:
“The whole point of having a digital assistant is to have it do stuff for you.
You’re supposed to boss it around. […] [B]ut if we’re going to live in a
world in which we’re ordering our machines around so casually, why do so
many of them have to have women’s names?” (LaFrance 2016). The
worry behind this question seems to be that the way we treat digital assis-
tants gendered as female will not leave our analog interpersonal relation-
ships and the way we conceive of ourselves as men and women unaffected.
In this regard, it could be argued in analogy with Kant in his Metaphysics
of Morals that even if the AI cannot be harmed by our behavior and that
we cannot violate any moral duty to the AI, we may fail with regard to it
(Kant 1797: 237f., Doctrine of the Elements of Ethics, Book I, §17). Our
behavior would thereby reveal shortcomings of our moral character and
our propensity to act wrongly when it comes to humans as well.
The broader critical twist of this second reason to bother amounts to
the AI being the user’s property. For, as mentioned earlier, even if the user
granted the AI its own identity and choice, the fact that one’s beloved
would at the same time be and remain one’s property would simply make
it impossible to treat the relationship as one of love between equals. This
obviously holds from the perspective of the owned AI but also from the
perspective of the dominant human. Hence, this—at least for now—
unavoidable fundamental asymmetry in the case of love between a human
and an AI would essentially make it impossible to realize fully either of the
three characterizations of romantic love described earlier in terms of the
included symmetry between the lovers. Hence, even if the human lover
wanted to have a fully symmetric relationship and meet the AI on an equal
footing, this is simply impossible. Consequently, we stick with our conclu-
sion that such a loving relationship cannot be considered worthwhile pur-
suing—assuming the human person does not pursue it for merely selfish
reasons to begin with.

What About Male AIs?

Finally, the considerations in the previous paragraphs give rise to yet
another objection: in the section “A Feminist Reading of Her,” we briefly
mentioned that gender stereotyping and sexism are not only detrimental
for women but also for men—and persons identifying with any gender, for
that matter. And while we claimed to say something general about loving
288 A. KLONSCHINSKI AND M. KÜHLER

relationships between humans and AIs, we have focused on AIs gendered

as female. While it may be true that AIs are predominantly gendered as
female both in science fiction and when it comes to digital assistants, it is
of course possible that, in the future, more and more AIs will be gendered
male so that we should say something about how our analysis applies to
AIs gendered as male, too.9
When imagining a loving relationship between a currently realistic
“male” AI and a human, the considerations presented in the section
“Implications for Loving Relationships” are equally valid: due to asym-
metries in the relationship, it cannot be considered a loving relationship
worth pursuing on any of the presented accounts. A difference arises,
however, as to the nature of the asymmetries: while with human beings,
the person gendered as female is usually subordinated, a relationship with
a “male” AI would probably subordinate the party marked as male. Such
relationships would thus not perpetuate and reinforce the existing sexist
social order and may even help to erode it. Yet, the AI may also be gen-
dered as male in the traditional sense and could thus reinforce sexist pat-
terns, as would be the case if, for instance, the male digital fitness trainer
yells at its owner to run faster or if the GPS system gives him or her direc-
tions with a deep authoritative voice. Since these gender stereotypes limit
freedom and opportunities for all, AIs should not be gendered as in this
sense traditionally male or female in the first place.

13.5   Conclusion
To sum up, we used the movie her as an example for formulating a feminist
ethical critique of loving relationships between humans and AIs in general.
We assumed that, within the constraints of actual technological possibili-
ties, any real AI would resemble Samantha’s “personality” at the begin-
ning of the movie, thus inviting the respective feminist critique. Due to
the asymmetries involved in relationships of love between humans and
AIs, we argued that, on any characterization of romantic love depicted
here, such love would not be worth pursuing. Finally, even though real

9
While examples of a woman falling in love with a robot gendered as male are certainly not
as prominent in science fiction, there are a few exceptions. In the Swedish series Real Humans
a woman falls in love with her good-looking Fitness-Hubot and in an episode of Star Trek:
The Next Generation (“In Theory”) a female officer falls in love with the Android Data (who
writes a dedicated love-subroutine in his programming specifically for her).
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 289

AIs could not be harmed, as they are not persons, such relationships,
would they become regular social practice, would also invite moral criti-
cism due to their likely detrimental social consequences. This is especially
true when it comes to AIs gendered as female.10

References
Anders, Charlie Jane. 2015. From Metropolis To Ex Machina: Why Are So Many
Robots Female? io9.
Aristotle EN. Nicomachean Ethics. Translated by W. D. Ross. The Internet
Classics Archive.
Becker, Julia C., and Chris G. Sibley. 2015. Sexism. In Handbook of Prejudice,
Stereotyping, and Discrimination, ed. Todd D. Nelson, 2nd ed., 315–336.
New York: Psychology Press.
Bogost, Ian. 2018. Sorry, Alexa Is Not a Feminist. The Atlantic. January 24.
Corliss, Richard. 2013. Spike Jonze’s her. Falling in love with the IT girl.
Review. Time.
Costa, Pedro, and Luísa Ribas. 2019. AI Becomes Her: Discussing Gender and
Artificial Intelligence. Technoetic Arts 17: 171–193. https://doi.org/10.1386/
tear\emph000141.
Dovidio, John F., Miles Hewstone, Peter Glick, and Victoria M. Esses, eds. 2010.
The Sage Handbook of Prejudice, Stereotyping and Discrimination. London:
Sage Publications.
Doyle, Sady. 2013. ‘Her’ Is Really More About ‘Him.’ In These Times,
December 20.
Fisher, Mark. 1990. Personal Love. London: Duckworth.
Frankfurt, Harry G. 1994. Autonomy, Necessity, and Love. In Necessity, Volition,
and Love, ed. Harry G. Frankfurt, 129–141. Cambridge: Cambridge
University Press.
———. 1999. On Caring. In Necessity, Volition, and Love, ed. Harry G. Frankfurt,
155–180. Cambridge: Cambridge University Press.
———. 2004. The Reasons of Love. Princeton: Princeton University Press.
Friedman, Marilyn. 1998. Romantic Love and Personal Autonomy. Midwest
Studies In Philosophy 22: 162–181. https://doi.org/10.1111/j.1475-
­4975.1998.tb00336.x.
Glick, Peter, and Susan T. Fiske. 2001. An Ambivalent Alliance. Hostile and
Benevolent Sexism as Complementary Justifications for Gender Inequality. The
American Psychologist 56: 109–118.

10
We would like to thank Simon Cushing for comments and criticism to an earlier draft of
this chapter, which were a tremendous help to improve it.
290 A. KLONSCHINSKI AND M. KÜHLER

Glick, Peter, and Laurie A. Rudman. 2010. Sexism. In The Sage Handbook of
Prejudice, Stereotyping and Discrimination, ed. John F. Dovidio, Miles
Hewstone, Peter Glick, and Victoria M. Esses, 328–344. London: Sage
Publications.
Gregoratto, Federica. 2017. Why Love Kills: Power, Gender Dichotomy, and
Romantic Femicide. Hypatia 32: 135–151. https://doi.org/10.1111/
hypa.12308.
Habermas, Jürgen. 1991. Vom pragmatischen, ethischen und moralischen
Gebrauch der praktischen Vernunft. In In Erläuterungen zur Diskursethik,
100–118. Frankfurt am Main: Suhrkamp.
Haines, Elizabeth L., Kay Deaux, and Nicole Lofaro. 2016. The Times They Are
a-Changing … or Are They Not? A Comparison of Gender Stereotypes,
1983–2014. Psychology of Women Quarterly 40. SAGE Publications Inc:
353–363. https://doi.org/10.1177/0361684316634081.
Hauskeller, Michael. 2017. Automatic Sweethearts for Transhumanists. In Robot
Sex. Social and Ethical Implications, ed. John Danaher and Neil McArthur,
203–217. Cambridge, MA: The MIT Press.
Helm, Bennett W. 2017. Love. In The Stanford Encyclopedia of Philosophy, ed.
Edward N. Zalta, Fall 2017. Metaphysics Research Lab, Stanford University.
Hoffmann, Magdalena. 2014. What Relationship Structure Tells Us About Love.
In Love and Its Objects. What Can We Care For? ed. Christian Maurer, Tony
Milligan, and Kamila Pacovská, 192–208. Basingstoke: Palgrave Macmillan.
Jollimore, Troy. 2015. “This Endless Space Between the Words”: The Limits of
Love in Spike Jonze’s Her. Midwest Studies in Philosophy 39: 120–143. https://
doi.org/10.1111/misp.12039.
Kant, Immanuel. 1797. The Metaphysics of Morals. Revised. Cambridge: Cambridge
University Press, 1996.
Kimmel, Michael. 2013. Angry White Men. American Masculinity at the End of an
Era. New York: Nation Books.
Krebs, Angelika. 2014. Between I and Thou—On the Dialogical Nature of Love.
In Love and Its Objects. What Can We Care For? ed. Christian Maurer, Tony
Milligan, and Kamila Pacovská, 7–24. Basingstoke: Palgrave Macmillan.
———. 2015. Zwischen Ich und Du: Eine dialogische Philosophie der Liebe.
Berlin: Suhrkamp.
Kühler, Michael. 2011. Love as Union vs. Personal Autonomy? In Love on Trial:
Adjusting and Assigning Relationships, ed. Nadine Farghaly and Corazón
T. Toralba, 93–104. Oxford: Inter-Disciplinary Press.
———. 2014. Loving Persons. Activity and Passivity in Romantic Love. In Love
and Its Objects. What Can We Care For? ed. Christian Maurer, Tony Milligan,
and Kamila Pacovská, 41–55. Basingstoke: Palgrave Macmillan.
———. 2020. Love and Conflicts Between Identity-Forming Values. In
International Handbook of Love: Transcultural and Transdisciplinary
 13 ROMANTIC LOVE BETWEEN HUMANS AND AIS: A FEMINIST ETHICAL… 291

Perspectives, ed. Claude-Hélène Mayer and Elisabeth Vanderheiden. Dordrecht:

Springer, 423–438. https://doi.org/10.1007/978-­3-­030-­45996-­3.
———. 2021. Love as Union and Political Liberalism. In Love, Justice, and
Autonomy. Philosophical Perspectives, Routledge Studies in Ethics and Moral
Theory, ed. Rachel Fedock, Michael Kühler, and Raja Rosenhagen. New York:
Routledge, 289–306.
LaFrance, Adrienne. 2016. Why Do So Many Digital Assistants Have Feminine
Names? The Atlantic. March 30.
Levin, Ira. 1972. The Stepford Wives. New York: Random House.
Levy, David. 2008. Love and Sex with Robots. The Evolution of Human-Robot
Relationships. New York: Harper Perennial.
Loh, Janina. 2019. Können Menschen Liebe machen? Eine inklusive Perspektive
auf Liebe, Freundschaft und Sex mit nichtmenschlichen Wesen. praefaktisch.
Manne, Kate. 2017. Down Girl. The Logic of Misogyny. Oxford: Oxford
University Press.
Merino, Noël. 2004. The Problem with “We”: Rethinking Joint Identity in
Romantic Love. Journal of Social Philosophy 35: 123–132.
Nozick, Robert. 1990. Love’s Bond. In The Examined Life. Philosophical
Meditations, ed. Robert Nozick, 68–86. New York: Simon & Schuster.
Nyholm, Sven, and Lily Eva Frank. 2017. From Sex Robots to Love Robots: Is
Mutual Love with a Robot Possible? In Robot Sex. Social and Ethical
Implications, ed. John Danaher and Neil McArthur, 219–243. Cambridge,
MA: The MIT Press.
Oshana, Marina, ed. 2014. Personal Autonomy and Social Oppression. New York:
Routledge.
Ricœur, Paul. 1995. Oneself as Another. Translated by Kathleen Blamey. Chicago:
University of Chicago Press.
Rorty, Amelie Oksenberg. 1987. The Historicity of Psychological Attitudes: Love
Is Not Love Which Alters Not When It Alteration Finds. Midwest Studies in
Philosophy 10: 399–412.
Roth, Abraham Sesshu. 2017. Shared Agency. In The Stanford Encyclopedia of
Philosophy, ed. Edward N. Zalta, Summer 2017. Metaphysics Research Lab,
Stanford University.
Soble, Alan. 1997. Union, Autonomy, and Concern. In Love Analyzed, ed. Roger
E. Lamb, 65–92. Boulder: Westview Press.
Solomon, Robert C. 1994. About Love. Reinventing Romance for Our Times.
Indianapolis: Hackett Pub. Co., Reprint 2006.
Steele, Chandra. 2018. The Real Reason Voice Assistants Are Female (and Why It
Matters). PCMag UK. January 4.
Sullins, John P. 2012. Robots, Love, and Sex: The Ethics of Building a Love
Machine. IEEE Transactions on Affective Computing 3: 398–409. https://doi.
org/10.1109/T-­AFFC.2012.31.
292 A. KLONSCHINSKI AND M. KÜHLER

v. Wedelstaedt, Almut Kristine. 2020. Gelungener Sex. Zeitschrift für Praktische

Philosophie 7: 103–132. https://doi.org/10.22613/zfpp/7.1.4.
Valian, Virginia. 1998. Why So Slow? The Advancement of Women. Cambridge,
MA: MIT Press.
Wilson, John. 1995. Love between Equals. A Philosophical Study of Love and Sexual
Relationships. Houndmills: Palgrave Macmillan.
 CHAPTER 14

Patriotism and Nationalism as Two Distinct

Ways of Loving One’s Country

Maria Ioannou, Martijn Boot, Ryan Wittingslow,

and Adriana Mattos

Love for a country has come to be linked with two concepts: patriotism
and nationalism. The distinction between these two concepts has been a
matter of controversy. In this chapter, we argue that one way of thinking
about and distinguishing between patriotism and nationalism is via the
very concept of love. More narrowly, we argue that love in patriotism is
similar to filial love (love for one’s parents), whereas love in nationalism
resembles intense passionate love. Our aim is twofold: (a) to get a clearer
picture of the different types of love that are involved in the conceptions
of patriotism and nationalism that we are interested in, and (b) to show
that the kind of love involved in our conception of patriotism can be
harmless, while the kind of love associated with the relevant conception of
nationalism can be dangerous and easily involves “bad faith,” a deceptive
faith in the superior goodness of one’s country.

M. Ioannou (*) • M. Boot • R. Wittingslow • A. Mattos

University of Groningen, Groningen, The Netherlands
e-mail: m.ioannou@rug.nl; m.g.j.boot@rug.nl; r.m.wittingslow@rug.nl;
a.mattos@rug.nl

© The Author(s), under exclusive license to Springer Nature 293

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8_14
294 M. IOANNOU ET AL.

We build up our argumentation as follows: in Part I we set out to pro-

vide grounds allowing us to treat both patriotism and nationalism as quali-
tatively distinct concepts before showing, in Part II, that one way of
distinguishing patriotism and nationalism is via the types of love they
implicate. In Part II we attempt to draw the links between patriotism and
filial love on the one hand and nationalism and intense passionate love on
the other. Our reasoning for Parts I and II draws from literature across the
following disciplines: philosophy, psychology, and biology.

14.1   Are Patriotism and Nationalism

Distinct Constructs?
Before we begin, let us consider our definitions, beginning with patrio-
tism. We define patriotism as entailing a love of one’s country, a sense of
personal identification with one’s country, and a special concern for its
well-being. This will be further developed in a later section.
What, then of “nationalism”? This poses a greater challenge. Both
patriotism and nationalism are bound by a love of one’s country; this
implies perhaps that nationalism is either the same as patriotism or a sub-
class of patriotism. And yet, nationalism and patriotism are treated as if
they are qualitatively distinct: there exists a general attitude that patriotism
is the principled, “good” kind of love for a country and nationalism the
nativist and “bad” kind of love.
Unfortunately, there is no clear consensus in the philosophical litera-
ture about how we can make this distinction. Indeed (and not for lack of
trying), given the multiplicity of uses, contested origins, and historical and
philosophical inheritances of both patriotism and nationalism, it is difficult
to isolate a feature or set of features that (a) consistently differentiated
nationalism from patriotism and (b) is capable of capturing nationalism in
its various forms. We found ourselves, in short, without a coherent and
universally agreeable concept of nationalism.
The more we thought about it, the less inclined we became to think
that rendering a coherent and universally agreeable concept of nationalism
is even possible: the idea is too broad, too contested, and too fraught with
historical contingency. Consequently, we came to the conclusion that,
contra general use, nationalism is not one single thing. Instead, it denotes
a cluster of overlapping phenomena, ideas, and attitudes that possess some
degree of family resemblance.
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 295

Naturally, family resemblances do not offer quite the same critical trac-
tion as concepts; absent structural integrity or genuine predictive power,
they are a little hard to nail down, let alone employ in a non-trivial way.
With that in mind, this chapter does not concern the family resemblance
between nationalisms: trying to force a family resemblance to do serious
work for us seemed an impossible task. Instead, and consequently, this
chapter deals with one particular articulation—one particular conception—
of nationalism. This conception of nationalism is the nationalism that
found its most potent, and most tragic, expression in Europe in the late
nineteenth to mid-twentieth century: the nationalism that served as a
foundational condition for World Wars I and II.
We selected this conception of nationalism for two reasons. First, it is
the conception of nationalism with which, we imagine, most readers
would be familiar. Even non-specialists possess some relevant facts about
the role of World War I in reforging European nation-states along linguis-
tic, ethnic, and cultural lines, or are to some degree familiar with how the
expansionist and genocidal policies of the Nazis were grounded in a par-
ticular conception of the German nation. We cannot assume this degree of
tacit familiarity with other conceptions of nationalism.
Second, we selected this particular conception of nationalism because,
unlike many other conceptions of nationalism, the shared etiology and
philosophical commitments of late nineteenth- to mid-twentieth-century
European nationalism are clear. At the risk of generalizing, we can say that
late nineteenth- to mid-twentieth-century European nationalism is typi-
fied by a set of self-same commitments about who can and cannot be
considered part of a given nation. Usually identified as emerging from the
Romanticism of Johann Gottfried Herder, who famously argued against
the universalizability of human needs, this conception of nationalism is
premised upon the claim that what makes someone French, or German, or
British, or whatever, is not simply a matter of contingency. Instead, what
makes someone part of a given nation is a shared ontology, whether that
ontology be linguistic, genealogical, cultural, genetic, or otherwise.
However, it would be a mistake to think that late nineteenth- to mid-­
twentieth-­century European nationalism is identical with Herder’s nation-
alism. Whilst Herder is an important rung in the intellectual genealogy of
our particular conception of nationalism, Herder was careful to argue that
the incomparability of different national groups means that it is simply
impossible for one national group to be better than any other.
Unfortunately, however, history had other ideas. As Isaiah Berlin argues,
296 M. IOANNOU ET AL.

the violent exigencies of eighteenth- and nineteenth-century geopolitics

turned Herder’s plea for autonomy into something both bloody and con-
ceptually incoherent: an “embittered and aggressive nationalist self-asser-
tion” that saw the political preferencing of possessed ontologies at the
expense of other, more distant ontologies (Berlin 1972: 17).
How, then, do these ontological assumptions manifest? We already
know all about the rise of nationalism on a historical scale. However, while
historians make sense of this shift on the scale of armies and cities and poli-
ties, for a more fine-grained understanding of how these intellectual shifts
affected (and affect) individual human behavior, we turn to psychological
literature.
Patriotism and nationalism became a topic of study for (social) psy-
chologists following World War II. Social psychology, as we will show
next, generally distinguishes between the two by treating them as two dif-
ferent types of national sentiments, and social-psychological research has
thus far predominantly focused on mapping these differences
(Sapounzis 2008).
Shortly after World War II, Theodor Adorno set out to study the phe-
nomenon of fascism as it manifested in the behavior of individuals. Given
the prominence of psychoanalysis at the time, Adorno and colleagues
departed from the assumption that there must be a personality trait that is
shaped by the nature of the relationship between children and parents,
and which explains what they regarded to be an abnormal behavior/ phe-
nomenon: the endorsement of fascist ideology. They found this trait in
what they dubbed the “authoritarian personality” (Adorno 1950).
According to them, the authoritarian personality described by a con-
stellation of traits—extreme respect and obedience to authority, obsession
with status and hierarchy, and the tendency to displace anger to weaker
individuals—predisposes the individual to ethnocentrism: “a tendency in
the individual to be ‘ethnically centered’, to be rigid in his acceptance of
the ‘culturally alike’, and in his rejection of the ‘unlike’” (Adorno 1950,
p. 102). One of the subscales of ethnocentrism Adorno and colleagues
constructed they labeled “pseudo-patriotism” and operationalized it as
uncritical conformity to prevailing societal norms and rejection of other
nations or outgroups. They understood “true patriotism” to simply be
love for a country: a benign, positive, beneficial, and even necessary attach-
ment to the nation.
Adorno and colleagues demonstrated using empirical data that pseudo-­
patriotism was positively correlated with authoritarian personality as well
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 297

as with outgroup derogation (more anti-Black and anti-Semitic sentiments

in the U.S.). What they did not provide was empirical evidence for the
distinction between pseudo-patriotism and true patriotism.
In search for an empirical distinction between the two terms, Kosterman
and Feshbach (1987) largely adopted the definitions of Adorno and his
colleagues for patriotism (true patriotism) and nationalism (pseudo-­
patriotism) and examined the extent to which the two could be conceived
as independent concepts. For this they created a measurement of patriotic
and nationalistic attitudes to statistically determine whether these two
types of attitudes tapped onto two distinct factors (patriotism and
nationalism).
Kosterman and Feshbach (1987) defined patriotism as love for and
pride in one’s nation and measured it using statements such as “I am emo-
tionally attached to my country” and “The fact that I am American is an
important part of my identity.” They construed nationalism as the percep-
tion of national superiority and orientation toward national dominance
and measured it using items such as “It is really important that the U.S. be
number one in whatever it does”; “The first duty of every young American
is to honor the national American history and heritage.” Their American
student participant scores on the two measurements indicated that the
two types of attitudes did indeed load on two different factors (thus rep-
resenting two different constructs) but that these two constructs were not
entirely independent from each other in the sense that they were signifi-
cantly correlated: the more participants endorsed nationalistic attitudes
the more they also endorsed patriotic attitudes. As Feshbach (1994)
remarked, however, even though related to patriotism, nationalism’s dis-
tinguishing feature remained to be that it “entails feelings of national
superiority, of competitiveness with other nations, and of the importance
of power over other nations” (p. 281).
Reviewing the existing psychological literature on the topic, Hanson
and O’Dwyer (2019) provided a useful way of thinking about the differ-
ences between patriotism and nationalism. They present nationalism and
patriotism as two distinct ways in which individuals can attach themselves
to their nation. They identify three elements on which these two ways of
relating may differ: affect, membership, and relations. Affect refers to
one’s attitudes and/or emotions toward their country; membership is
about who belongs to the country; and relation is about how individuals
relate to figures or institutions of power in their country.
298 M. IOANNOU ET AL.

With regard to affect, a patriot’s positive affect for her country is self-­
referential, meaning that it originates from the mere fact that it is her
country. Meanwhile, for a nationalist, her positive affect stems from an
almost exclusively downward comparison between her own and other
countries. As far as membership is concerned, a nationalist and a patriot
would give different answers to who belongs to their ingroup. A national-
ist would set the bar to be ethno-cultural cohesion (ingroup members are
those with whom she shares common heritage), whereas a patriot will
answer using civic lines (ingroup members are those with whom she shares
common purpose and goals). Finally, in regard to relations with the coun-
try’s authorities, a nationalist would subscribe to the dictum “my country,
right or wrong.” She would hold an unquestionable positive evaluation of
the country’s institutions of power and would be intolerant to any criti-
cism directed to them. A patriot, by contrast, would be open to question-
ing and criticizing the wrongdoings of her country’s institutions of power.
As noted, one defining characteristic distinguishing patriotism and
nationalism according to social-psychological literature is that the first is
self-referential and the latter relies heavily on intergroup comparison in
search of superiority. Patriotism is described by what social psychologists
would call “ingroup love” and nationalism by both ingroup love and out-
group derogation: “outgroup hate” (Brewer 1999). What has been a mat-
ter of controversy in psychology, however, is the extent to which ingroup
love is bound to give rise to outgroup derogation or outgroup hate
(Hewstone et al. 2002). The question in other words is whether patrio-
tism paves the way to nationalism.
More recent studies in the crossroads of biology and psychology read-
dressed this old question, asking whether ingroup love (or ingroup favor-
itism) is part and parcel with outgroup derogation or not. To do so they
investigated the effects of oxytocin on the propensity of people to behave
either more pro-ingroup (demonstrating ingroup preference and/or
favoritism) or more anti-outgroup (demonstrating outgroup derogation).
Oxytocin was chosen because of its repeatedly demonstrated role in pro-
moting trust and cooperation (only) among ingroup members (Kosfeld
et al. 2005).
With this knowledge about the properties of oxytocin at hand, research-
ers in the Netherlands conducted a series of experiments asking their male
Dutch participants to self-administer oxytocin or placebo before submit-
ting them to behavioral tests aiming at measuring ingroup favoritism ver-
sus outgroup derogation (de Dreu et al. 2011). An example of such a
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 299

behavioral test was an intergroup version of Philippa Foot’s original trolley

dilemma where the participants could pull the lever to kill one to save five
(Foot 1967). The twist was that the oxytocin and the placebo participants
were further split into two more conditions: in the ingroup target condi-
tion they could pull the lever to kill a fellow ingroup member (Dutch),
and in the outgroup target condition they could choose to pull the lever
knowing that they would kill an outgroup member (Arab or German).
They then assessed whether oxytocin participants by comparison to the
placebo participants were more likely to exhibit ingroup favoritism (by
being less likely to pull the lever to kill the Dutch person to save five
unidentified individuals less often) or outgroup derogation (by being
more likely to pull the lever to kill an outgroup member to save five
unidentified individuals more often). What they found is that participants
in the oxytocin condition exhibited more ingroup favoring behavior than
their placebo counterparts but not more outgroup derogatory behavior.
Similar results were found with Chinese female and male participants.
Those inhaling oxytocin had higher likability ratings for Chinese people
than did the placebo participants. Their likability of stimuli associated with
other Asian countries (Taiwan, South Korea, and Japan), however, was
unaffected by oxytocin (Ma et al. 2014).
These results suggest that ingroup favoritism and outgroup derogation
(or ingroup love and outgroup hate) are distinct processes for they are
differentially affected by a neurotransmitter that serves to heighten trust
and cooperativeness with ingroup members. If ingroup favoritism and
outgroup derogation are indeed processes that distinguish patriotism and
nationalism, then these studies provide further support to that patriotism
and nationalism are not one and the same and that nationalism does not
just follow from patriotism.
What we did thus far was to show that there are reasons allowing us to
think of patriotism and nationalism as two different concepts. We also
explained why, especially in the case of nationalism, we are focusing on
one conception of nationalism, that is nationalism as it manifested in nine-
teenth- and twentieth-century Europe. This conception of nationalism
aligns well with the way nationalism was studied in the social sciences and
social and political psychology in particular. In the remainder of this chap-
ter we will attempt to show another way in which patriotism and national-
ism differ and that is via the type of love they invoke. To show that the
conceptions of patriotism and nationalism we are interested in implicate
300 M. IOANNOU ET AL.

two different types of love, we first attempt to draw the analogy between
patriotism and filial love and then the one between nationalism and intense
passionate love.

14.2   Analogies Between Patriotism and Filial

Love and Nationalism and Intense Passionate Love
Patriotism and Filial Love
The scholarly literature gives different descriptions of patriotism. However,
these descriptions have the following characteristics in common: love of
one’s country, a sense of personal identification with one’s country, and a
special concern for its well-being. In this section, we will discuss similari-
ties between patriotism and filial love to clarify the type of love that is
involved in the case of patriotism.
The analogy between patriotism and filial love is suggested by the word
patriotism itself, which is deduced from the Latin pater (father). Some
thinkers regard patriotism as a vice. One of them is Leo Tolstoy. Tolstoy
(1967) rejects patriotism because he thinks that it is opposed to the ideal
of morality, which is “the recognition and brotherhood of all men.”
Patriotism replaces this by “the recognition of one state and nationality as
predominating over all the others.” According to Tolstoy, patriotism lacks
any moral foundation.
Also, Keller (2005) believes, for other reasons, that patriotism is a vice
rather than a virtue. He claims that there is a connection between patrio-
tism and bad faith. In this context, “bad faith” means deceptive faith in the
superior goodness of one’s country, a faith that persists even if there is
objective counter-evidence. Keller further argues that a patriot cannot
maintain the love for her country if she discovers that her country has seri-
ous shortcomings.
We will show, with the help of an investigation of its similarities with
filial love, that patriotism, characterized by the three features mentioned
earlier, need not be morally reprehensible. It may be compatible with an
impartial and objective outlook and it may avoid bad faith. Besides, we will
argue that patriotism need not be given up if the patria (homeland)
appears to be less perfect than supposed.
We will make use of John Rawls’s discussion of the three stages of moral
development in a well-ordered liberal-democratic society. This Rawlsian
account supports the similarities between filial love and patriotism and
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 301

shows that patriotism need not cause a distorted view of the qualities of
one’s country, just like filial love need not cause a distorted view of the
qualities of one’s parents.

Analogy Between Patriotism and Filial Love

In this section, we will first focus on filial love and its analogy with patrio-
tism, before we turn in the next section to a kind of reciprocity, which is
present both in the relation between parents-children, and citizens-fellow
citizens.
Because people are neither free to choose their parents nor free to
choose their native country, neither the love of children for their parents
nor the love of citizens for their country is the result of a deliberate deci-
sion. Both kinds of love are mainly given, due to the fact that they concern
our roots: our parents are the persons from whom we originated and our
country is the place where we grew up.
The intimate relation between children and their parents and the con-
nection between citizens and their fellow-countrymen and native soil
explain another common characteristic of the two types of love: the pre-
paredness to fight for the well-being of one’s parents and for the interests
of one’s mother country, even at the cost of sacrifices.
The patriot may explicate her love by mentioning valuable characteris-
tics of her country. Just as a daughter may mention positive characteristics
of her father (“intelligent, kind, fair, supportive, courageous”), a patriot
may describe the qualities of her country (“freedom, tolerance, open-
ness”). However, just as positive characteristics of her father do not wholly,
not even mainly, determine why his daughter loves him, positive charac-
teristics of her mother country do not wholly, not even mainly, determine
why the patriot loves it. Again, both kinds of love are grounded merely in
the fact that they concern one’s parents and one’s native country, while
their specific characteristics play a supportive role.
If we ask why you love your father, no justification is needed beyond
your reply that “he is my father.” Similarly, if we ask you why you love
your country, you need not provide another justification than that it is the
country where you were born. Keller (2005) agrees with the former claim,
but disagrees with the latter. According to him, a patriot will and should
base her love on valuable characteristics of her country.
It is true that a patriot can refer to valuable features of her country.
However, what she will deny is that these characteristics are the main
302 M. IOANNOU ET AL.

reason for, or cause of, loving her country. In a way that is analogous to
explaining filial love, she will rather emphasize that she feels love for her
mother country because it is the place where she grew up in relation with
fellow-countrymen and because her mother country’s history, culture,
and language shaped her identity.
When the patriot is pressured to explain what is so lovable about her
country, she will be able to point out its good qualities. However, just as
she may continue to love her father while recognizing some negative fea-
tures of his character and behavior, she may continue to love her country
without closing her eyes for its shortcomings and possible reprehensible
conduct. A country embodies a plurality of characteristics, of which some
are better and others are worse. If some features are imperfect or bad, they
need not be a reason for decreasing love.
Keller, by contrast, thinks that a patriot cannot uphold her love if she is
confronted with evidence that her country is less good than supposed. He
thinks that a patriot will be inclined to deny this evidence or to deceive
herself by interpreting the evidence in a biased way so that she does not
have to draw such a conclusion. This is what Keller (2005) calls “bad
faith.” He argues that it is difficult for a patriot to assess the evidence
impartially, precisely because counter-evidence for the goodness of her
country would be a reason to decrease or give up her love.
However, in the next section, we shall argue that persons, who have
been brought up and educated well in a liberal-democratic constitutional
state, are capable of combining patriotism with an impartial outlook and
avoiding bad faith.

Patriotism and Impartiality

We will discuss the three stages of moral development put forward by
John Rawls in his A Theory of Justice to show that patriotism is compatible
with a capacity to be impartial. This Rawlsian account supports the anal-
ogy between filial love and patriotism.1
Rawls (1999) discusses the idea of a well-ordered liberal-democratic
society. It is a society that is designed to advance the good of all its mem-
bers and is regulated by a fair system of cooperation, which is acceptable

1
The concept of patriotism is not discussed in Rawls’s theory. However, it plays an impor-
tant role in the transition from the second stage of the “morality of association” to the third
stage of the “morality of principle” (see Callan 2006).
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 303

to all. Fair cooperation requires reciprocity: everybody has to do their part

and everybody will benefit, according to the rules of fairness. Reciprocity
lies between impartiality and mutual advantage.
This fairness is part of the public political culture of a liberal-democratic
constitutional society. A fundamental characteristic of fairness is that mor-
ally arbitrary factors should not play a role in the distribution of advan-
tages and burdens across the citizens of the society. Examples of morally
arbitrary factors are social class, race, gender, origin, and place of resi-
dence. To the extent that these factors are considered as being irrelevant
for principles of justice, the fairness that is part of a liberal-democratic
society is impartial and unbiased.
It is to be expected that people who grow up and live in a public politi-
cal culture of fair social cooperation, in which impartiality plays a central
role, will develop a sense of justice and impartiality in moral questions and
will be capable of avoiding biased and distorted views with respect to
themselves and others.
As Rawls describes, this capability comes into existence during three
stages of moral development, which, interestingly, concerns both the rela-
tion of children toward their parents and the relation of citizens toward
their fellow citizens.
During the first stage—“the morality of authority”—children are sub-
ject to the authority of their parents. Parental love means not only being
concerned with the wants and needs of the child, but also affirming her
sense of self-respect. Parental love is expressed, among others, by the par-
ents’ care for the child, enjoyment of her company, and encouragement of
her attempts to gain control over her development. Although a young
child has the potentiality for love, her actions are at first motivated by
instincts and desires and guided by rational self-interest. When the child
notices that the parents love her, she develops love and trust for her par-
ents in return (Rawls 1999, p. 405). The child slowly acquires skills and a
sense of competence. She understands her success and enjoyment are par-
tially the result of her following the example and norms of her parents. She
feels gratitude toward her parents for taking care of her.
The “morality of association” is the second stage of moral develop-
ment. Associations are, for instance, the school, the neighborhood, the
company, the family itself, and the national community as a whole. In
society, an individual fulfills a role in different associations (Rawls 1999,
p. 405). As the child grows up, older members of the association explain
the expectations and virtues appropriate for fulfilling her role. Each
304 M. IOANNOU ET AL.

association has its moral ideals, defined in ways that are appropriate for the
respective role. So, as an adult, the person learns how to behave and act
according to the virtues suitable for her role in her occupation and as a
member of society.
In a manner that is comparable to the relation of the child toward her
parents, a citizen develops a love of country when she experiences that her
nation cares for her, encourages her education, protects her self-respect,
and offers her opportunities to flourish. She understands that her success
and well-being are partially the result of her following the rules of society.
She feels gratitude toward her country, so she wants to cooperate and do
her part as the societal rules require.
As Rawls (1999) writes:

Thus we may suppose that there is a morality of association in which the

members of society view one another as equals, as friends and associates,
joined together in a system of cooperation known to be for the advantage of
all and governed by a common conception of justice. The content of this
morality is characterized by the cooperative virtues: those of justice and fair-
ness, fidelity and trust, integrity and impartiality. The typical vices are grasp-
ingness and unfairness, dishonesty and deceit, prejudice and bias. (p. 405)

The third stage of moral development is the “morality of principles.” A

citizen who succeeds in following the moral standards that apply to her in
her different roles in various associations naturally comes to an under-
standing of the principles of justice that rule her society. Her motive for
complying with the principles of justice follows from the established
friendships, attitudes of love, trust, fellow feeling for others, and a concern
for her society. The process whereby a citizen becomes attached to the
principles themselves, and her wish for acting justly, is analogous to the
way she came to follow the injunctions of her parents as a result of her love
for them.
In a well-ordered society, the citizens’ capacity for being impartial leads
them in the end to master the principles of justice and understand the
values they protect and the way they are to everyone’s advantage (Rawls
1999: 414–415).
Through complex forms of the morality of association, a citizen devel-
oped attachments to many different individuals and communities. Her
grasp of the morality of principles makes her understand that her care
needs to be extended to all citizens.
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 305

A possible objection has still to be mentioned. Rawls speaks about a

social contract of fair cooperation between citizens who belong to the
same society. However, patriotism is criticized for possible biased views,
not with respect to fellow citizens but with respect to one’s own country
compared to others.
If citizens have a sense of justice and are capable of being fair and
impartial to fellow citizens who belong to other associations, then it is not
a very big step to extend these virtues to a supranational level (Rawls
1999: 414–415).
A patriot need not view her own country as the best. Her adherence to
the principle of impartiality will enable her to assess evidence fairly and to
avoid “bad faith.” Although she feels love for her country, her grasp of the
morality of principles also makes her aware that, from an impartial point of
view, the place where she was born is contingent and morally arbitrary. She
therefore understands that also other countries and their citizens deserve
fair treatment and respect.
Note that this version of patriotic love is not uncritical. Uncritical patri-
otic love applies to more extreme versions of patriotism and not necessar-
ily to the kind of patriotism under consideration. Earlier we defined
according to three “central characteristics” the patriotism in which we are
interested. We described the analogy between filial love and this version of
patriotism. In this section we explained why the patriotic love under con-
sideration is compatible with an impartial and objective outlook. This is
also supported by the social-psychological literature as we discussed it in
Part I: our patriot appears to be open to questioning and criticizing the
wrongdoings of their country’s institutions of power.
In this section we argued that John Rawls’ explication of the three
stages of moral development supports the thesis that patriotism need not
cause a distorted view of the qualities of one’s country, just like filial love
need not cause a distorted view of the qualities of one’s parents.
A person that has gone through Rawls’ three phases of moral develop-
ment is less susceptible to a biased and uncritical attitude toward her fam-
ily and is capable of having respect for other people and capable of treating
them fairly.
In an analogous way, a patriot who has gone through Rawls’ three
phases of moral development is less susceptible to a biased and uncritical
attitude toward her own country. In this way a patriot’s special concern for
(and her giving priority to) the well-being of her own country does not
entail a negatively biased or unfair treatment of other countries.
306 M. IOANNOU ET AL.

Nationalism and Intense Passionate Love

The conception we use of nationalism has the following key characteris-
tics: love for one’s country that is not self-referential but is fueled by the
(downward) comparison of one’s country with others, love for one’s
country is uncritical leading to an unquestionable positive evaluation of
the country and its institutions, coupled with intolerance to any criticism
directed at them.
There are many definitions offered for intense passionate love in litera-
ture. The religious theoretician C.S. Lewis uses the word eros to describe
one of the four loves (Lewis 1960). Eros, according to Lewis, is that state
of “being in love” (p. 131), described by a “delighted preoccupation with
the beloved” and which combines “strength” “sweetness” and “terror”
(p. 133).
Sociologist John Alan Lee identifies three styles of romantic love, one
of which is eros. Eros according to him is an intensely emotional experience
described by a very powerful attraction to the beloved, a strong preoccu-
pation with them, and the strong wish for the relationship to remain
exclusive (Lee 1973).
Biological anthropologist Helen Fisher breaks down romantic love into
three phases: lust, attraction, and attachment (Fisher 2014). The first
phase, lust, is merely physical and the individual is governed by their basic
drives. The last phase (attachment) is related to a wish for a more lasting
nurturing commitment. Attraction is the phase of love that is described by
passion (hence passionate love). In intense passionate love (or attraction),
according to Fisher, the lover regards their beloved as special and unique,
their attention is exclusively focused on their beloved, and they overem-
phasize their beloved’s better traits while downplaying or overlooking
their flaws.
These definitions share some key characteristics all of which are con-
tained in Fisher’s (2014) description of intense passionate love. We side
with Fisher in that description, which leads us to define intense passion-
ate love as love toward a person someone is romantically interested in
that is described by (a) intense and powerful attraction to the object of
love, (b) a strong preoccupation with the object of love, and (c) idealiza-
tion of the object of love. We adopt the strong view that for a type of love
to be categorized as intense passionate love it has to meet all three of the
aforementioned characteristics. Naturally not all people go through this
phase of love and not all instances of passionate love need to be described
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 307

by these three characteristics. Also, since we are drawing upon theories

built on empirical studies, we can only safely talk about inclinations: a
person experiencing intense passionate love is strongly inclined to behave
in the aforementioned ways.

Analogy Between Nationalism and Intense Passionate Love

To draw the analogy between nationalism and intense passionate love, we
show how the two resemble each other in the following three ways: (i) in
both nationalism and passionate love there is a (desperate) attempt to
cling on to one’s object of love; (ii) both nationalism and passionate love
render one susceptible to “bad faith” via compromising impartiality; and
(iii) in both nationalism and intense passionate love, love for the love-­
object is experienced as a “high” that may be bound to be short-lived
(especially in the case if intense passionate love) but that is also likely to be
re-triggered time and again.

( i) Clinging on to the Object of Love

We have already argued that in filial love (as well as in patriotism) there is
absence of choice: we are not free to choose either our parents or our
native country. They are both given. Even though in romantic love there
may be an element of choice, or that for true or authentic love real choice
is required (Kane 1998); this is not the case for intense passionate love as
we defined it earlier. Intense passionate love is often accompanied by the
conviction that there is no real choice over who we fall in love with, that
we are simply destined to be with our beloved. As Lewis (1960) put it:
“half the love songs and half the love poems in the world will tell you that
the Beloved is your fate or destiny, no more your choice than a thunder-
bolt […]” (p. 126).
The difference though between filial love and intense passionate love is
that in intense passionate love there is no guarantee for a permanent
attachment to our beloved. This is to say that while there is lack of choice
in both passionate love and filial love, this lack of choice has different emo-
tional consequences in these two types of love. The key difference origi-
nates from the fact that in passionate love this lack of choice is coupled
with a feeling of lack of agency (lack of control over the destiny of one’s
love) which then breeds fear, insecurity, and strong dependence. Findings
from empirical studies attest to this psychological profile of a person
308 M. IOANNOU ET AL.

experiencing intense passionate love. Fisher and her colleagues (2016)

have found in neuroimaging studies that those individuals reporting being
possessed by passionate love behaved in ways that resembled the behavior
of both a person rejected in love and an addict. In their own words:

[…] men and women who are passionately in love and/or rejected in love
show the basic symptoms of substance-related and gambling addiction listed
in the Diagnostic and Statistical Manual of Mental Disorders-5, including
craving, mood modification, tolerance, emotional and physical dependence
and withdrawal. (Fisher et al. 2016: 1)

We argue that precisely because of the fear that one may lose one’s
beloved, or because one is subconsciously aware of the transient nature of
this state of love, one is bound to cling onto one’s beloved, to hold them
tight, so as to preserve what one has, at all costs. What “at all costs” may
translate to is crimes in the name of love. As Fisher et al. (2016) put it,
even in its harmless form, intense passionate love “is associated with
intense craving and can impel the lover to believe, say and do dangerous
and inappropriate things” (p. 2).
A nationalist now may, too, believe that she just happens to be Dutch,
Brazilian, Australian, or Cypriot and that there was no choice involved in
that. But, a nationalist also holds a strong view of what being Dutch,
Brazilian, Australian, or Cypriot means or should mean. They are “in
love” with a particular version of their country. A nationalist for example
would have a strong and inflexible view about who belongs to their
ingroup. For them, ingroup members are exclusively those with whom
they share common heritage (Li and Brewer 2004). They will also hold
the strong view that their country possesses properties that place it in a
position of superiority by comparison to other countries (Kosterman and
Feshbach 1987). But as is the case with the beloved in passionate love
where there is little guarantee that the beloved is here to stay, the perma-
nency of the particular version of the country a nationalist is in love with
is not to be guaranteed either.
Hence, just like passionate lovers, nationalists will do anything to cling
onto their beloved or rather the version of their beloved they are infatu-
ated with. For nationalists, this means investing themselves in keeping
their object of love—their country—intact, unaltered. A nationalist will do
so by constantly ensuring that their country is the same country they are
in love with, either by continuously comparing it with other countries in a
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 309

light that favors it or by keeping their country and its identity free from
“malevolent” and “contaminating” sources. It is no coincidence that the
(anti-) immigration discourse of nationalists is imbued with what in litera-
ture is known as realistic and symbolic threats, that is, threats to harming
or compromising the country’s physical integrity or threats to the coun-
try’s meaning system: a threat to all those things like religion values and
belief systems that make their country their country (Kadianaki et al.
2018). Immigrants, thus, are seen as a threat to the country’s physical and
symbolic integrity or in other words a threat to the version of the country
a nationalist is in love with.
A nationalist, similarly with a person in intense passionate love, would
do anything to cling onto the version of their country they are in love with
including conceding to the killing of people who threaten their country’s
physical and symbolic integrity. Feshbach (1990), for example, found that
nationalists and not patriots were more supportive of nuclear armament
policies and were more willing (for their country) to go to war.

( ii) Bad Faith

In passionate love one’s beloved is idealized and stands out in positive
ways. What is driving the passion is the conviction that there is something
truly unique about one’s beloved. In fact, passion in intense passionate
love is compromised when one starts discovering that one’s partner is
more average, that is, less special than one thought them to be (Fisher
2014). In the same way, love for a country in nationalism is fueled by the
firm belief that one’s country is truly unique.
If someone pointed out to a person experiencing passionate love that
their beloved has certain bad qualities, the chances are that this person,
preoccupied and fused as they would be in this stage of romantic love,
would readily and uncritically dismiss these judgments. Functional mag-
netic resonance imaging (fMRI) studies elucidating neural correlates of
intense passionate love have found that there are certain areas of the brain
that are either overactivated or deactivated when someone experiences
passionate love. For example, areas important to the brain’s reward system
are overactivated, whereas areas associated with negative emotions and
areas associated with higher level cognitive functions such as judgment are
deactivated (Zeki 2007). This, according to de Boer and colleagues,
explains the observation that people experiencing passionate love are
unable to truly judge their beloved’s qualities (De Boer et al. 2012).
310 M. IOANNOU ET AL.

Similarly, with a person experiencing passionate love, a nationalist

would react with defensiveness to someone pointing out to them qualities
of their country that are not good. If they are confronted with evidence
that their country is less good than supposed, then the reaction is not one
of impartiality as we would expect from a patriot but a reaction similar to
the one of a lover who is immersed in passionate love. We would not
expect a nationalist—as we would not expect a person in passionate love—
to engage critically with this information, but rather to dismiss it.

( iii) Love “Highs”

Subjective experience and now research evidence show that intense pas-
sionate love is experienced as a “high,” described by increased energy
(hypomania), elation when things go well, despair when not, separation
anxiety when apart, and general anxiety about how things will turn out
(Fisher 2014). All in all, this is a fascinating, but largely unsustainable,
state of being which explains the relatively short duration of intense pas-
sionate love (De Boer et al. 2012).
Even though intense passionate love is doomed to die out sooner or
later, there are two interesting exceptions to this one-time “high” general
rule. The first is that while the intensity of passionate love may ease out,
there are cases of couples who remain “in love” despite the passage of
time. There are, for example, fMRI investigations comparing brain activa-
tion of men and women who had recently fallen in love and of men and
women in their 50s and 60s in long-term marriages who reported that
they still felt the “high” of early stage romantic love. These studies show
an overlap in brain areas activated in the two groups of participants—par-
ticularly areas involved in the reward system—when looking at a photo-
graph of their beloved while being scanned (Fisher 2014). A key difference
though between those who recently fall in love and those who still experi-
ence the highs of love but are in longer-term relationships is the amount of
stress levels experienced. Cortisol levels of individuals who had fallen in
love within 6 months compared with their cortisol levels 12–24 months
later show that the initial increase registered in cortisol levels in the first 6
months of being in love was not observed later, when the relationship had
entered a more long-term stage (Marazziti and Canale 2004). These show
that the feeling of being “in love,” which is a key characteristic of passion-
ate love, may sedate without, however, getting completely lost.
The second exception is that the “high” experienced in passionate love
may be re-triggered when love is endangered. The “symptoms” of
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 311

passionate love: intense preoccupation with the beloved, emotional and

physical dependence on them, craving for them, and so on may, for exam-
ple, return when the lover experiences rejection from the beloved. Fisher
and colleagues call this a relapse. In their own words: “[…] long after the
relationship is over, events, people, places, songs, and/or other external
cues associated with their abandoning sweetheart can trigger memories
and initiate renewed craving, obsessive thinking and/or compulsive call-
ing, writing or showing up in hopes of rekindling the romance—despite
what they suspect may lead to adverse consequences” (Fisher et al. 2016:
2). This relapse, of course, refers to a negative “high” or, as Fisher and
colleagues put it, a negative addiction.
Nationalism (and patriotism) are typically seen as orientations or ways
of loving that are relatively stable and difficult to change (Kosterman and
Feshbach 1987). There is, however, reason to believe that these orienta-
tions must fluctuate across situations and that they are aroused under well-­
defined conditions. Druckman (1998) discusses an extensive range of
studies showing that this is especially the case for nationalistic attitudes
which will be more pronounced—they will spike—in competitive inter-
group situations. This is why nationalism, as history also has shown, will
surge, will reach a “high,” when the version of the country a nationalist is
in love with is threatened. We could think, for example, of how nationalist
sentiments spike during economic crises, times of mass migration move-
ments, or how they have surged during the recent refugee crisis in Europe.
We do not think there is a very good reason for nationalistic sentiments
to remain elevated when the country’s existence, superiority, and purity
are not under some kind of threat. In such circumstances of low competi-
tion, we would expect that the sentiments of a nationalist will resemble the
sentiments of a person in a long-term and secure relationship with their
beloved. Once though the country’s existence, superiority, and purity are
threatened, then that will be similar to endangered love in the romantic
love parallel. This perceived threat will, as we noted earlier, rekindle the
“high”: one’s preoccupation, obsession, dependence on one’s object of
love, out of fear of losing it. Like Fisher et al. (2016), we deem this “high”
that is induced by fear of losing one’s beloved, to be a negative “high,” to
be a reaction to a strong aversion that can lead someone to a bad place, by
which we mean committing crimes in the name of love.
312 M. IOANNOU ET AL.

14.3   Conclusion
In this chapter we illustrated why we think patriotism and nationalism (or
at least the conceptions of them that are of interest to us) to both be
instances of love for a country that invoke, however, two different types of
love. To build our argument we first explained why we are interested in
the particular conceptions of patriotism and nationalism and why there are
reasons to think of patriotism and nationalism as qualitatively distinct con-
cepts. We then proceeded to show that love for a country in patriotism is
analogous to the love of children toward their parents, whereas love for a
country in nationalism is analogous to intense passionate love toward a
lover. In drawing these analogies, we also showed how patriotism, just like
filial love, may guard itself from “bad faith” by showing that it is possible
for a patriot as it is possible in filial love, to love without biased judgments.
Via the analogy between nationalism and intense passionate love, we
showed how love for a country in nationalism, even though fulsome and
fervent, may be dangerous in the same way passionate love may end up
being dangerous. We argued that love for a country in nationalism, just
like passionate love, is most dangerous when the object of love is threat-
ened, thus explaining why nationalism is particularly likely to reach a
mindless and reckless “high” in cases of competition when the existence,
the superiority, or the purity of the love object—their country—are seen
to be at risk.

References
Adorno, Theodor W. 1950. The Authoritarian Personality. New York, NY: Harper.
Berlin, Isaiah. 1972. The Bent Twig: On the Rise of Nationalism. Foreign Affairs
51: 11–20.
Brewer, Marilynn B. 1999. The Psychology of Prejudice: Ingroup Love or
Outgroup Hate? Journal of Social Issues 55: 429–444.
Callan, Eamon. 2006. Love, Idolatry, and Patriotism. Social Theory and Practice
32: 525–546.
De Boer, Antina, Erin M. van Buel, and Gert J. Ter Horst. 2012. Love Is More
Than Just a Kiss: A Neurobiological Perspective on Love and Affection.
Neuroscience 201: 114–124.
De Dreu, Carsten K.W., Lindred L. Greer, Van Kleef, A. Gerben, Shaul Shalvi, and
Michel J.J. Handgraaf. 2011. Oxytocin Promotes Human Ethnocentrism.
PNAS 108: 1262–1266.
 14 PATRIOTISM AND NATIONALISM AS TWO DISTINCT WAYS OF LOVING… 313

Druckman, Daniel. 1998. Nationalism, Patriotism, and Group Loyalty: A Social

Psychological Perspective. Mershon International Studies Review 38: 43–68.
Feshbach, Seymour. 1990. Psychology, Human Violence, and the Search for
Peace: Issues in Science and Social Values. Journal of Social Issues 46: 183–198.
———. 1994. Nationalism, Patriotism and Aggression: A Clarification of
Functional Differences. In Aggressive Behaviour: Current Perspectives, Plenum
Series in Social/Clinical Psychology, ed. L.R. Huesmann, 275–291. New York:
Plenum Press.
Fisher, Helen E. 2014. The Tyranny of Love: Love Addiction—An Anthropologist’s
View. In Behavioral Addictions: Criteria, Evidence and Treatment, ed. Laura
Curtiss Feder and Ken Rosenberg, 237–265. London: Elsevier Press.
Fisher, Helen E., Xiaomeng Xu, Arthur Aron, and Lucy L. Brown. 2016. Intense,
Passionate, Romantic Love: A Natural Addiction? How the Fields That
Investigate Romance and Substance Abuse Can Inform Each Other. Frontiers
in Psychology 7: 867. https://doi.org/10.3389/fpsyg.2016.00687.
Foot, Philippa. 1967. The Problem of Abortion and the Doctrine of the Double
Effect. Oxford Review 5: 5–15.
Hanson, Kristin, and Emma O’Dwyer. 2019. Patriotism and Nationalism, Left
and Right: A Q-Methodology Study of American National Identity. Political
Psychology 40: 777–795.
Hewstone, Miles, Martin Rubin, and Hazel Willis. 2002. Intergroup bias. Annual
Review of Psychology 53: 575–604.
Kadianaki, Irini, Maria Avraamidou, Maria Ioannou, and Elisavet Panagiotou.
2018. Understanding Media Debate Around Migration: The Relation Between
Favorable and Unfavorable Representations of Migration in the Greek Cypriot
Press. Peace and Conflict: Journal of Peace Psychology 24: 407–415.
Kane, Robert. 1998. The Significance of Free Will. New York: Oxford
University Press.
Keller, Simon. 2005. Patriotism as Bad Faith. Ethics 115: 563–592.
Kosfeld, Michael, Markus Heinrichs, Paul Zak, Urs Fischbacher, and Erst Fehr.
2005. Oxytocin Increases Trust in Humans. Nature 435: 673–676.
Kosterman, Rick, and Seymour Feshbach. 1987. Towards a Measure of Patriotic
and Nationalistic Attitudes. Political Psychology 10: 257–274.
Lee, John A. 1973. Colours of Love: An Exploration of the Ways of Loving. Toronto:
New Press.
Lewis, Clive Staples. 1960. The Four Loves. New York: Harcourt, Brace.
Li, Qiong, and Marilynn B. Brewer. 2004. What Does It Mean to Be American?
Patriotism, Nationalism and American Identity After 9/11. Political Psychology
25: 727–739.
Ma, Xiaole, Lizhu Luo, Yayuan Geng, Weihua Zhao, Qiong Zhang, and Keith
M. Kendrick. 2014. Oxytocin Increases Liking for a Country’s People and
314 M. IOANNOU ET AL.

National Flag But Not for Other Cultural Symbols or Consumer Products.
Frontiers in Behavioural Science 8: 266.
Marazziti, Donatella, and Domenico Canale. 2004. Hormonal Changes When
Falling in Love. Psychoneuroendocrinology 29: 931–936.
Rawls, John. 1999. A Theory of Justice, Revised edition. Cambridge, MA: The
Belknap Press of Harvard University Press.
Sapounzis, Antonis. 2008. Towards a Critical Social Psychological Account of
National Sentiments: Patriotism and Nationalism Revisited. Social and
Personality Psychology Compass 2: 34–50.
Tolstoy, Leo. 1967. Tolstoy’s Writings on Civil Disobedience and Nonviolence.
New York: New American Library.
Zeki, Semir. 2007. The Neurobiology of Love. FEBS Letters 581: 2575–2579.
 Index1

A Autonomy, 14, 21, 29, 159, 274, 281,

Abramson, Kate, 63, 174, 175, 177, 283, 284, 285n7, 296
186, 186n7, 195n1, 200n5,
228n10, 259n18
Adams, Robert, 195n1, 200n5 B
Adorno, Theodor, 296 Badhwar, Neera Kapur, 5, 26, 29,
Affect-program theorists, 135 31, 33, 57
Agape, 31, 107n7, 209 Bagley, Benjamin, 199n3
Agent Bagnoli, Carla, 9, 44n6, 65n1, 80, 223n5
free, 13, 15, 167 Baker, Lynn, 75
moral (see Moral) Baron, Marcia, 27n2, 217n2
Alvarez, Maria, 179, 184 Beatrice, 9, 69–73, 76, 80, 83
Ameliorative project, 8, 64, 67 Becker, Julia, 272
Anders, Charlie Jane, 286 Beliefs, 17, 47, 57, 76, 142, 143n34,
Animals, non-human, 19 145, 152, 153, 156, 180n5, 182,
Aristotle, 28, 35, 50n15, 136n23, 186, 186n7, 191, 196, 210,
197, 277 257, 281
Arpaly, Nomy, 204n10 false, 137, 184
Attention, 41 optimistically biased, 75
Audi, Robert, 143n35 as reasons, 184
Austen, Jane, 57 stereotypes, 272

1
Note: Page numbers followed by ‘n’ refer to notes.

© The Author(s), under exclusive license to Springer Nature 315

Switzerland AG 2021
S. Cushing (ed.), New Philosophical Essays on Love and Loving,
https://doi.org/10.1007/978-3-030-72324-8
316 INDEX
Berlin, Isaiah, 296
Blameworthiness, 92, 156, 160, 162,
162n9, 162n10, 165n13
Bogost, Ian, 273
Bortolotti, Lisa, 9, 75–78
Bragg, Billy, 16
Brewer, Marilynn, 308
Brink, David O., 241, 253n14
Broackes, Justin, 49
Brogaard, Berit, 6, 35n15, 36, 195n1,
200n5, 227
Brown, Robert, 253n14,
259n18, 260n19
Burns, Gregory, 20, 244, 244n5, 245,
245n7, 246n8, 247n10, 263
Buunk, Bram, 76
C Delaney, Neil, 174, 182
Callan, Eamon, 302n1
Canale, Domenico, 310
Categorical Imperative, 4
Cavell, Stanley, 210
Chalmers, David, 133n18
Chang, Ruth, 173n3
Character, 280, 302
change of, 176
moral, 22, 78, 135, 174, 177, 197,
223n6, 234, 287
traits of, 7, 50, 84, 177, 188,
216n1, 223, 231n13, 254, 280
Children
love between parents and, 5, 22, 58,
67, 76, 121, 161, 177, 197,
204n11, 206, 207n13, 227n9,
230, 233n15, 277, 293,
296, 300–302
Clarke, Bridget, 44n7, 84
Clausen, Ginger, 195n1, 199n3
Cocking, Dean, 155
Consent, 113, 284
Contempt, 136
Corliss, Richard, 272
Costa, Pedro, 286
Cuypers, Stefaan, 152n4, 162n10
D
Damasio, Antonio, 128n7, 135n21
Dancy, Jonathan, 18, 203, 203n9,
208, 209n16
Dante Alighieri, 9, 69, 71–73,
76, 80, 83
Darwall, Stephen, 223n5
Davenport, J., 127n4
Davidson, Donald, 185
de Boer, Antina, 309
de Dreu, Carsten, 298
de Sousa, Ronald, 56, 127n3, 134n20,
140n30, 172n1, 191, 196n2
Deonna, Julien, 128n7, 172n1, 173,
182, 187
Deontic properties, 18, 112, 203
Determinism, 82, 164, 164n11, 167
Dignity of persons, 31, 65, 67,
73, 257n16
contrast with price, 4, 154, 224n7
Dillon, Robin, 30
Disposition, 17, 20, 50, 127, 127n5,
134, 146, 189, 224, 230,
231n13, 252, 253, 258,
259, 265
Double, Richard, 157n6
Dovidio, John F., 272
Doyle, Sady, 272, 275
Driver, Julia, 207, 208
Drowning wife (Fried/Williams
example), 2, 18, 43n4, 160,
215–226, 221n4
Druckman, Daniel, 311
Drummond, John J., 223n5
Duty, 3, 6, 10, 14, 25, 27, 44, 89,
97–99, 118, 158, 160, 287

E Forgiveness, 96, 139, 160, 164, 201
Ebels-Duggan, Kyla, 27n7,
113n11, 201n7
Ecclesiastes, 10
Ehman, Robert, 260n19
Ekman, P., 128n7, 136n23
Emery, Robert, 75
Emotion, 7, 12, 56, 107, 127, 128,
132–138, 135n21
Empathy, 20, 245, 246
Empirical persona, 32, 33, 67, 68, 71
The End of the Affair, 12, 125, 146
Enoch, David, 95n6
Enticing Reasons View, 202–208
Eros, 45, 107n7, 306
Evaluative properties, 18,
111–112, 203
Existentialism, 10, 81, 96
Ex Machina, 286
Exploitation, 92
F Fromm, Erich, 22
Fantasies, 9, 47, 69–72, 74–79
Fear, 133, 173, 174, 181, 204, 206,
233, 307
Fehr, Beverley, 130n12
Feminine mystique, 74
Feminism, 74, 270–275, 284, 288
Ferreira, J.M., 127n4
Feshbach, Seymour, 297, 308,
309, 311
Fidelity, 138, 199, 304
Fisher, Helen, 306, 308–311
Fisher, Mark, 21, 107n8, 281
Fiske, Susan, 273
Fitting response, 173, 182, 188,
190, 233
Flourishing, 4, 119, 153, 230,
277, 278
Fogal, Daniel, 179, 191n9
Foot, Philippa, 299
INDEX 317
Formula of Humanity, Kant’s, 29
Foundationalism, 143, 143n35
Fowers, Blaine, 75
Frank, Lily, 276
Frankfurt, Harry, 10, 15, 18, 19, 21,
63, 65, 98, 127n4, 153, 155,
156, 158, 196n2, 204n11, 215,
217n2, 220–222, 227–230,
231n13, 237, 253n14, 254n15,
257, 257n17, 260n19, 277
Franklin-Hall, Andrew, 20, 253n14,
256–259, 265
Freud, Sigmund, 3, 31
Fried, Charles, 2, 18, 216
Friedan, Betty, 74
Friedman, Marilyn, 282
Friendship, 5, 14, 25, 27–32, 35, 70,
87, 90, 93, 129, 137, 155, 159,
216n1, 232, 234, 304
Kantian, 28
sexual, 130, 139
Frowe, Helen, 103
Fungibility, 29, 140n30, 186, 262
G
Gender, 8, 79, 85, 274, 286, 287, 303
roles, 273
stereotype, 21, 270, 274, 281, 287
Gerschwin, George and Ira, 94n5
Gert, Joshua, 200n4
Gertler, Brie, 133n18
Giles, James, 253n14
Glick, Peter, 272, 273
Good, the, 45n8, 55, 59, 82, 210
Grau, Christopher, 187,
189n8, 260n19
Green, O.H., 153, 153n5, 254n15
Greene, Graham, 12, 125, 126n2, 129n9
Greenspan, Patricia, 200n4
318 INDEX
Gregoratto, Federica, 272, 274, 276
Grenberg, Jeanine, 65n1
Grief, 154, 215, 241, 243n3, 279
Griffiths, Paul, 135n22
H Jordan, Jessy, 47n11
Habermas, Jürgen, 271
Haines, Elizabeth, 272
Han, Yongming, 189
Hanson, Kristin, 297
Harcourt, Edward, 68
Haslanger, Sally, 8, 64
Hauskeller, Michael, 276n4, 286
Hegel, G.W.F., 81
Helm, Bennett, 106n6, 107n7, 116n13,
141n32, 233n16, 253n14, 257,
257n17, 260n19, 277n5
Her (film), 21, 269
Herder, Johann Gottfried, 295
Hoffman, Eric, 259n18
Hoffmann, Magdalena, 276
hooks, bell, 64, 71, 74
Hopwood, Mark, 7, 43, 45
Hornsby, Jennifer, 184n6
Hurka, Thomas, 20, 195n1, 199n3,
253–256, 258, 259, 265
Hyman, John, 184n6
I Kraut, Robert, 176, 260n19
Illusions, positive, 75
Ingroup, 298, 299, 308
Inherent value, 19, 44, 65, 216,
216n1, 223–225, 223n6, 224n7,
225n8, 227, 229, 233, 233n16,
235, 237, 238, 257n16
J LaFrance, Adrienne, 273, 287
Jaworska, Agnieszka, 20, 253n14,
256–259, 265
Jesus, 60
John, Eileen, 83
Jollimore, Troy, 56n23, 63, 175, 176,
195n1, 198, 200n5, 217n2,
253n14, 259n18,
261n19, 262n20
Jonze, Spike, 21, 269, 271
Jus in bello, 104
Just War Theory, 10, 102–106
K
Kadianaki, Irini, 309
Kagan, Shelly, 200n5
Kant, Immanuel, 3–5, 14, 22, 25,
27–29, 32, 44, 65, 66n2, 81,
154, 206, 217, 224n7, 287
Keller, Simon, 39n1, 63, 174, 182,
195n1, 217n2, 235, 300, 301
Kennett, Jeanette, 155
Kimmel, Michael, 275
Knowledge, moral, 51
Kolodny, Niko, 10, 15, 18, 19, 63,
93n4, 153, 154, 176, 177, 184,
185, 189, 195n1, 199n3, 200n5,
207, 221, 228n10, 231n13,
232–234, 253n14, 257, 260n19
Korsgaard, Christine, 173n3
Kosfeld, Michael, 298
Kosterman, Rick, 297, 308, 311
Krebs, Angelika, 21, 279
Krenwinkel, Patricia, 131n14
Kroeker, Esther Engels, 195n1,
202, 207n12
L
LaFollette, Hugh, 253n14, 260n19
Lamb, Roger, 158, 162n8, 260n19
Landrum, Ty, 260n19
Lazarus, R.S., 128n7, 135n21, 136n23

Lee, John Alan, 306
Leite, Adam, 63, 174, 175, 177, 186,
186n7, 195n1, 200n5,
228n10, 259n18
Levy, David, 270, 278, 279, 283, 286
Lewis, C.S., 306, 307
Li, Qiong, 308
Liao, S. Matthew, 207n13
Little, Margaret Olivia, 200n4,
202, 203n9
The Little Prince, 10, 88
Loh, Janina, 285
Lopez-Cantero, Pilar, 8, 68, 69, 71
Lord, Errol, 184, 204n10
Lovable, 17, 59, 59n25, 67, 151, 159,
163, 174
Love
as actively attending, 72, 77
as arresting awareness, 44
Christian, 60n26
commitments of, 138–144
dual account of, 226
epistemic role of, 42
ersatz, 14, 157
filial (see children)
as historical, 159
incomplete, 254
ingroup, 23, 298
meaningful, 20, 21, 65, 70, 71, 75,
84, 263, 264
moral (Kantian), 26, 27, 32
“natural,” 27
parental (see children)
passionate, 294, 300–312
“pathological,” 27n2, 28, 44
as perception, 42
phenomenology of, 56, 107, 128n6
as a practice, 83
as a psychological condition, 245,
249, 260
as really looking, 65
requirements of, 158–159, 227
INDEX 319
romantic, 26, 35, 36, 58, 87, 99,
129, 130n12, 133, 138, 145,
153n5, 187, 208n14, 216n1,
234, 242, 247, 247n9, 269,
275, 277, 279, 281, 283,
306, 309
selectivity of, 7, 58, 67, 68, 140,
175, 178, 231, 231n13, 232
sexual, 58
sibling, 234
triangular theory of, 12, 127, 127n3
tripartite theory of, 20,
243, 261–265
as a union, 21, 107n8, 140, 270,
275, 276n3, 281–283, 285n7
unrequited, 10, 76, 99, 249
as valuing, 67
as a verb, 64, 72
as a virtue, 50
M
M and D
Iris Murdoch example, 6, 7, 9, 42,
47, 49, 49n13, 50, 52, 59,
72, 80, 82
Ma, Xiaole, 299
Macnamara, Coleen, 200n4, 202
Manne, Kate, 78, 273, 273n1, 275
Marazziti, Donatella, 310
Mason, Elinor, 217n2
Matthew, 60
McDowell, John, 49n14, 184n6
McLean, Mark, 50n15
Mele, Alfred, 13, 120n15, 152, 156
Merino, Noël, 282
Merritt, Melissa McBay, 65n1
Millgram, Elijah, 68
Milligan, Tony, 65, 65n1, 66n2,
243n3, 243n4
Moral agent, 6, 10, 15, 26, 219, 235
Moral realism, 46
320 INDEX

Mulhall, Stephen, 48n12 Passion, 5, 12, 23, 27, 127–129,

Murdoch, Iris, 3, 6, 7, 9, 39–60, 64, 128n6, 133n19, 138, 172, 182,
66, 72, 80, 82 293, 306–307, 309
Murray, Sandra, 76 sexual, 131
Paton, H.J., 27n2
Patriotism, 22, 293–312
N Perception
Naar, Hichem, 57, 174, 177, 187, value neutral, 48
195n1, 198, 245 Pereboom, Derk, 14, 153, 163–168
Narrative, 68 Pettit, Philip, 158n7, 207n12
Nationalism, 22, 293–312 Phenomenal concepts, 133n18
Nebel, Jacob, 203n9 Phenomenology, 44n6, 99, 128,
Needs 190, 205
in a relationship, 90, 91 Philia, 107n7, 277
Neu, Jerome, 201n8 Pismenny, Arina, 261n19
New Testament, 5, 8, 60 Plato, 1, 40n2, 45, 50n15,
Noggle, Robert, 30 65n1, 210
Noller, Patricia, 253n14, 259n18 Pride and Prejudice, 57
Nonsubstitutability, problem of, 31, Prinz, Jesse, 128n7, 135,
176, 183, 186, 231n13, 233n16, 135n21, 261n19
260, 262, 264 Protasi, Sara, 177, 195n1
No Reasons View, 15–17, 189, 192,
196–200, 196n2, 205, 210
Nozick, Robert, 21, 260n19, 281, 282 R
Nyholm, Sven, 276 Radical reversal cases, 155, 159,
162, 166
Rawls, John, 300
O theory of moral
Oakley, Justin, 159 development, 302–305
Obsession, 119, 247, 251, 252n13, Raz, Joseph, 173n3, 184, 203n9
253, 254, 296, 311 Real Humans, 288n9
O’Dwyer, Emma, 297 Reasons
One Bad Day (thought agent-relative, 198
experiment), 13, 152 different types of, 179
Oshana, Marina, 284 enticing, 18, 203n9, 209n16, 210
Othello, 2, 41, 53, 57 exclusionary, 198, 209
Oxytocin, 23, 298 explanatory, 180, 197
maximum, 236
non-insistent, 200n5
P normative, 181
Parfit, Derek, 184 responsivity to, 183
Partiality, 2, 5, 18, 25, 39n1, 206, special, 178, 216, 226, 234–237
219, 222, 224 warranting, 17, 199–202, 208

Relationship, 1, 8–10, 14, 15, 18, 20,
21, 26, 33, 36, 50, 53, 58, 63,
65, 70, 72, 75, 76, 93, 95, 97,
97n7, 99, 103, 106, 109, 110,
112, 118, 122, 128, 129, 131,
132n17, 137, 140, 151, 158,
159, 163, 174, 177, 182, 189,
198, 207, 208n14, 216n1,
220–222, 225, 227, 229–235,
237, 245, 248, 266, 269, 271,
273, 275–288, 306, 310
theory of love, 93n4, 178n4,
199n3, 232, 233n15, 257
Respect, 98
Kantian interpretation of, 4, 28, 44,
65, 205, 223, 226, 257n16
Responsibility
externalism (historicism), 13, 152
internalism (anti-historicism), 13,
152, 157, 157n6
prospective, 89, 93, 99
retrospective, 89, 92
second-order, 96
skepticism, 13, 163, 165
Ribas, Luísa, 286
Rich, Adrienne, 70, 79
Ricœur, Paul, 271
Robertson, Simon, 203n9
Romeo and Juliet, 129
Rorty, Amelie Oksenberg, 139n29,
187, 280
Roth, Abraham Sesshu, 279
Rudman, Laurie, 272
Rusbult, Caryl, 76
S Stocker, Michael, 10, 98
Safina, Carl, 20, 244, 246, 247n10,
251, 253, 264
Sapounzis, Antonis, 296
Scanlon, T.M., 184, 200n4
Schauber, Nancy, 56
INDEX 321
Schaubroeck, Katrien, 202, 207n12
Schroeder, Mark, 173n3
Schroeder, Timothy, 204n10
Science, 7, 81, 84
Science fiction, 278, 286, 288, 288n9
Seinfeld, 12
Sensen, Oliver, 27n2
Sentiments, 159–163, 186–189, 281
Setiya, Kieran, 16, 56, 175, 177, 178,
178n4, 190, 195n1, 198, 200n5,
201n8, 205
Sexual attraction, 127, 131, 133, 145
Sexual relations, 284n6
Shakespeare, William, 16, 31, 129,
130n11, 139
Shpall, Sam, 20, 243, 253n14,
259n18, 261–265
Sibley, Chris, 272
Smilansky, Saul, 167n17
Smith, Michael, 207
Smuts, Aaron, 63, 177, 189, 196n2,
201n8, 202, 245n6,
253n14, 261n19
Smyth, Nicholas, 217n2,
218, 219n3
Soble, Alan, 253n14, 261n19, 285n7
Solomon, Robert, 21, 96, 107n8,
141n31, 195n1, 197,
260n19, 281
Star Trek, 288n9
Steele, Chandra, 273, 286
The Stepford Wives, 278, 286
Stereotype, 21, 79, 270, 272, 273,
275, 281, 282, 287, 288
Sternberg, R.J., 12, 127, 129,
130n13, 131n15, 142
Strawson, Galen, 167n17
Sullins, John P., 270, 278, 283, 286
Supererogatory, 97, 112, 201
Swan Lake, 52
Sylvan, Kurt, 204n10
322 INDEX

T Wedgwood, Ralph, 204n10

Taking responsibility, 9, 87–99
Teasdale, J.D., 137n24
Teroni, Fabrice, 128n7, 172n1, 173,
182, 187
Thomas, Laurence, 155
Thought, one too many, 3,
18, 217–220
Tolstoy, Leo, 300
Trading up, 16, 175, 186, 198
V Williamson, Timothy, 184n6
Valian, Virginia, 272
Valuation, 19, 31, 32, 34, 154, 155,
222, 225, 226, 238
van den Eijnden, Regina, 76
Velleman, J. David, 3, 5, 7, 8, 15, 18,
19, 26, 29, 31–36, 43–45, 64–69,
71, 73, 116n13, 154, 162n8,
178, 178n4, 195n1, 198, 199n3,
201n6, 205, 207, 215, 217n2,
222–227, 229, 232, 237,
253n14, 257n16, 260n19
Virtues, 50, 60, 92
unity of, 53
Volition, 12, 127n4, 128, 143, 153,
228, 257, 277
Vow, marital, 10, 95–97, 138, 138n27
Weil, Simone, 60n26
Well-being, 94
White, Richard, 253n14,
254n15, 260n19
Whiting, Daniel, 200n4
Widdows, Heather, 78n5
Williams, Bernard, 2, 18, 25, 39,
43n4, 160, 216, 217, 219n3,
220, 222, 225n8
Williams-Beuren Syndrome, 20,
250, 250n11
Wilson, John, 284
Wittgenstein, Ludwig,
126n2, 210
Wolf, Naomi, 74
Wolf, Susan, 9, 53n19, 59n25, 92, 96,
217n2, 218, 228n11
Wollheim, Richard, 187
Wonderly, Monique, 245, 247n9, 252,
253n14, 254n15, 261n19
Wynne, Clive, 20, 244, 244n5, 248,
250, 250n11, 251, 253, 256,
259, 263, 265
Y
Yeats, W. B., 16

W Z
Wallace, R. Jay, 207n12 Zangwill, Nick, 9, 11, 63, 173, 196n2,
Watson, Gary, 157n6 253n14, 261n19
Wedelstaedt, Almut Kristine, 284n6 Zeki, Semir, 309